Labor Market

Highlights The Communist Party will hold its nineteenth National Congress on Oct. 18. This is the "midterm election" for President Xi Jinping, whose political capital will be replenished; Recent Chinese leaders have a greater impact in their second term than their first; Base case: Xi consolidates power while preserving a balance on the Politburo Standing Committee; Stay long Chinese equities versus emerging market peers. Feature China's Communist Party will hold the nineteenth National Party Congress on October 18-25. This is a critical "midterm" leadership reshuffle that will also mark the halfway point of General Secretary Xi Jinping's term in office. Investors around the world will watch closely to see what insight can be gained about the political trajectory of the world's second-largest economy. This report serves as a "primer" for readers to understand the party congress and its investment takeaways. Why Is The Party Congress Important? Because it rotates China's political leaders! Chart 1So Long To The 18th Central Committee China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer In a political system without popular representation, the rotation of personnel according to promotion and retirement is the only way to rejuvenate the policy process. The average rate of turnover on the Communist Party's Central Committee at each five-year congress has been 62%, which is a remarkably high rate (Chart 1). It reveals an underrated dynamism in Chinese politics. This leadership rotation also allows the top leader (Xi Jinping) to consolidate power by putting his supporters into key positions. This in turn alters the policymaking environment and the way in which China formulates policies and responds to external events. China has a "parallel" political system in which the ruling Communist Party operates alongside (and above) the state. Xi Jinping is "General Secretary" of the party, president of the People's Republic of China, and (not least) chairman of the Central Military Commission. The party maintains supremacy by independently controlling the state and the army. Since fall 2016, Xi has been dubbed the "core" of the Communist Party, putting him on a par with previous core leaders Mao Zedong, Deng Xiaoping and Jiang Zemin.1 The party's nearly 90 million members convene large congresses of about 2,000 members every five years to select the membership of the key decision-making bodies (Diagram 1), a practice known as "intra-party democracy."2 The key body is the Central Committee, which consists of about 200 full members and another 100-some alternative members. The Central Committee then "elects" the General Secretary, Political Bureau (a.k.a. "Politburo," the top 25 or so leaders) and Politburo Standing Committee (the "PSC," the top five-to-nine leaders) - though in reality the Politburo and the PSC are chosen through intense negotiations among the incumbent PSC and former leaders. Diagram 1National Party Congress Of The Communist Party Of China China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer The handful of men on the PSC are the chief decision-makers in China, often in league with the broader Politburo (and former PSC members who exercise some power through the back door). Most of the key personnel decisions will have been made before the Central Committee votes.3 Hence the current top leaders have a chance to put their loyalists and supporters in key positions, potentially improving the implementation of their agenda. The outgoing eighteenth Central Committee will meet for its last session on October 11, and then the nineteenth party congress will meet on October 18 to elect a new Central Committee. It will in turn ratify the new Politburo and PSC. At the beginning of the party congress, Xi Jinping will deliver a keynote political report on the state of the party and nation, reviewing the progress of the past five years and mapping out a vision for the next five. The party congress will also amend the Communist Party constitution.4 By the end of the week, the members of the new PSC will step out to meet the press together for the first time. Only later will the party's key decisions be incorporated by the state, i.e. China's central government, including key personnel appointments and policy initiatives. This will occur when the legislature, the National People's Congress ("NPC," not to be confused with party congress), convenes at its annual "Two Sessions" in early March 2018. Chart 2Bold Action Can Follow Midterm Congresses China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer Any NPC session following a five-year party congress carries more weight than usual not only because it approves of the party congress's leadership decisions but also because it kicks off major new policy initiatives. For instance, Premier Zhu Rongji was appointed to launch the "assault stage" of President Jiang Zemin's reforms of state-owned enterprise at the NPC in March 1998 (Chart 2). Similarly, Hu Jintao's Premier Wen Jiabao launched extensive administrative reforms at the NPC meeting in early 2008.5 How does a "midterm" party congress differ from others? Typically, in even-numbered years, the top two leaders change over, as with Xi Jinping and Premier Li Keqiang in 2012. These transitions are highly significant as they mark a leadership succession, a transfer of power to a new general secretary in a heavily centralized, authoritarian system that does not have a codified succession process. By contrast, in odd-numbered years like 2017, the Communist Party promotes, demotes, and retires a large number of other top leaders. Thus Xi Jinping's place is assured, and Li Keqiang's place is probably assured as well, but most likely the other five members of the PSC will be gone.6 This year's transition is also significant because the total turnover on the Central Committee is expected to be higher than usual (perhaps 70%) as a result of President Xi's aggressive anti-corruption campaign and other factors (see Chart 1 above).7 Leaders often spend the bulk of their first five years consolidating power and the second five years pushing forward their true policy agenda. Even President Hu Jintao, who failed to see his preferred social safety-net policies fully implemented, had a vastly more influential second term than first term in office: the 2007-12 period saw the 4 trillion RMB stimulus package to thwart the Global Recession. Moreover, Chinese leaders do not normally become "lame ducks" toward the end of their last term: Deng Xiaoping recommitted the country to pro-market reforms in 1992, after having stepped down as general secretary, while Jiang Zemin reached the height of his power at the end of his term in 2002, when he chose to hang onto the position of top military leader for two extra years. Many observers suspect that Xi Jinping will hold onto power beyond 2022. Bottom Line: The National Party Congress coincides with a sweeping rotation of the Chinese political elites, which is a critical way of ensuring that China, unlike a monarchy or personalized "dictatorship," has an orderly way of updating its policy-makers and (hopefully) policies. Midterm reshuffles allow top leaders to promote supporters and re-energize the implementation of their policy agenda. The past two Chinese leaders were more consequential in their second term than their first. How Is The Nineteenth Congress Unique? Chart 3Xi Jinping's Generation Taking Command China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer The most important change this year is the passing of a generation.8 China's political elites are classified into "leadership generations," with Mao Zedong symbolizing the first generation, Deng Xiaoping the second, Jiang Zemin the third, Hu Jintao the fourth, and Xi Jinping the fifth generation. The current reshuffle will see the following generational trends: The End of the Jiang Zemin Era: The key figures retiring on the PSC are those who were born before 1950 and put in place by Jiang Zemin. Thus in a very real sense, Jiang Zemin's influence is coming to a close (Chart 3).9 This generational shift is likely to force the retirement of 11 of the 25-member Politburo, and five of the seven PSC members (Table 1), as well as other major figures, such as the long-serving central bank Governor Zhou Xiaochuan. Table 1Chinese Leaders Set To Retire On Politburo And Politburo Standing Committee China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer Jiang-era leaders are defined by certain characteristics that are now fading. As Chart 4 demonstrates, these leaders came of age in the early, idealistic days of the Revolution, leading them to have a conservative streak in ideological matters. Yet they are well-known pragmatists in economic matters. They studied engineering and natural sciences in answer to the call for the young to develop the country's heavy industry. They tended to hail from capitalist-leaning coastal provinces, and often gained first-hand experience operating China's state-owned enterprises. This last point became especially important when they pioneered pro-market corporate reforms in the 1990s. By contrast, fewer of them served as government ministers on the State Council (China's cabinet) than subsequent generations. Chart 4Leadership Characteristics Of The Politburo Standing Committee China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer The Middle of the Hu Jintao Era: The passing of Jiang's cohort will necessarily give his successor Hu Jintao's cohort a boost in relative influence at the top levels. Hu's generation is marked by leaders who studied the "soft sciences" (like law and economics). Several of them (including Hu and Premier Wen Jiabao) have links with the politically liberal wing of the party. They have far less experience in the military or state-owned business, but are more likely to have governing experience in the central government and especially the provinces (Chart 4 above). This includes the interior provinces from which they often hail. They are thus highly attuned to the problem of maintaining social stability, arguably to the neglect of economic dynamism. Hu Jintao's influence may be underrated. Xi's administration has shown important continuities with Hu's, and Hu's followers are well positioned in the Central Committee, the Politburo, and the provincial governments (though not the current PSC). If Xi does not take decisive moves to replace some of Hu's acolytes on the PSC at the coming party congress, then Hu's men will likely outnumber Xi's on the PSC as they graduate up the ladder from the Politburo.10 A strong showing by Hu's faction could affect China's policy priorities, given that Xi showed different preferences from Hu in the first few years of his rule (Table 2). However, the factions do not maintain consistent policy platforms. The bottom line is that Hu's faction could act as more or less of a constraint on Xi regardless of what policies the latter pursues. Table 2Fiscal Priorities Of Recent Chinese Presidents China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer The True Beginning of the Xi Jinping Era: Xi's generation has yet to reveal its full character - the demographics of the new Central Committee will help determine it. So far it is a continuation of the trends above: more likely than not to come from interior than coastal provinces, to have studied the humanities, to have governed in the provinces or central ministries, and to lack military or business experience (Chart 4 above). The coming reshuffle could initiate a change in some of these trends, given some of Xi's revealed preferences, but that will not become clear until this fall.11 Xi is not stereotypical when it comes to China's political cycles: he consolidated power rapidly in his first term.12 The question, then, is whether Xi can continue to accrue power at the party congress, or whether his second term will become complicated by an infusion of Hu Jintao supporters into top party posts. Thus the success of Hu's supporters (particularly on the PSC) is the critical moving part that could determine the political constraints on Xi Jinping from 2017-22. Will Xi be able to arrange a favorable power-sharing agreement? Or will he go further and try to remove this political constraint entirely, even at the risk of political instability? The above points raise two critical questions: Will Chinese politics become more institutionalized? Investors should expect China to maintain a stridently informal political system. Rules and norms can and will be bent, but key principles will be upheld. In other words, the goal posts can be moved, but not too far. Going beyond certain limits would be destabilizing for China's political, institutional, and factional balances, and so far Xi has exhibited poise and the desire to maintain stability that is characteristic of post-1978 Chinese leaders.13 We think there is a low probability that Xi will overthrow all the norms of leadership selection and overturn the balance of power on the Politburo and PSC. If he does, it will raise alarms that he is setting up a new "cult of personality" like Mao, which could cause domestic economic and market instability. Rather, we expect him to modify the rules to maintain control of the PSC without excluding Hu Jintao's faction from power. Will Xi initiate the succession process for 2022? Some commentators suspect that Xi will use the party congress to pave the way for him to cling to power beyond 2022. Clearly Xi could retain the top military post and stay within recent precedent. But any hints at altering recent succession patterns, despite the fact that they are informal, are dangerous for investors in the long run because they raise deep uncertainty about the range of possibilities and political conflicts that could occur upon the actual change of power in 2022. Nevertheless, bear in mind the following points: The question of succession will not be resolved this October. If Xi plans to hang on beyond 2022, then he will continue amassing power and positioning loyalists over the next five years so that he will have full institutional support at the critical moment in 2022 - like Jiang Zemin did when he chose to hang onto the military chairmanship from 2002-04. Thus while Xi may lay some groundwork that makes political observers uneasy, the question will not be resolved either way this fall. Xi's tenure will be an ongoing topic for investors to monitor. Xi is already set to be the most powerful Chinese leader well into the 2020s. Xi's anti-corruption campaign is remarkable evidence of his strength as a ruler. Significantly, this campaign has focused on rooting out Jiang Zemin's influence. Yet Jiang stepped down way back in 2004! In other words, Jiang wielded massive influence between 2004 and 2017. Indeed, Xi's boldest move this year so far was to remove Sun Zhengcai, a Jiang acolyte. It stands to reason that, even if Hu Jintao's faction pulls off a relative victory this year, Xi Jinping's faction will likely be well positioned for a victory in 2022. And if Hu loses out this year, Xi's followers will be better positioned in 2027, as well as 2022. In short, market participants are unlikely to be able to tell the difference this October between (1) Xi getting a boost of political capital for his second term and (2) Xi getting such a big boost that he is on track to overstay his second term.14 Xi might intend to become a dictator and cling to power for longer, but all the market will know for certain is that he has maintained control of the PSC and his general policy framework will be more or less continuous, which is likely a relief in the near term. Finally, investors may not initially care if Xi seizes additional power at the expense of party norms and the succession process. A-shares sold off, but H-shares rallied, when Jiang Zemin decided not to step down entirely in 2002 (Chart 5). Russian stocks and the RUB/USD only fleetingly sold off when Vladimir Putin made clear his intention to return to the presidency yet again in 2011 (Chart 6). Chart 5Foreign Investors Cheered Jiang's Clinging To Power Foreign Investors Cheered Jiang's Clinging To Power Foreign Investors Cheered Jiang's Clinging To Power Chart 6Russian Investors Cheered Putin's Second Presidency Russian Investors Cheered Putin's Second Presidency Russian Investors Cheered Putin's Second Presidency While it is impossible to know whether markets will cheer any signs of "Papa Xi" doing away with term limits, it is bad for China's governance in the long run if Xi does not clearly begin grooming a successor with this fall's promotions. An heir-apparent for 2022 would reduce the risks of disruptive power-struggle and would impose a personal deadline on Xi Jinping's reform agenda. That is, a deadline above and beyond the 2020 deadline in the 13th Five Year Plan and the 2021 deadline for the 100th anniversary of the Communist Party's founding. That reform agenda, in turn, is essential for improving China's long-term productivity.15 Bottom Line: The Chinese political system is informal, which means that rules and norms can be bent without altering the underlying principles of balance among the key factions and stability of the regime and society as a whole. Our baseline scenario is a market-positive one: that Xi Jinping will win a victory at the party congress, but that he will not overthrow Hu Jintao's followers and abandon the "collective leadership" model, since that would destroy the overall balance of power and heighten domestic political risks. If Xi loses out to the Hu faction, then we would expect Chinese and China-exposed risk assets to sell off, at least initially. If Xi romps to total victory, excluding Hu's clique from power, we would fade any market rally. Such a development would heighten political risks for the foreseeable future. Investment Conclusions The prospect of a Xi-dominated, yet stable, PSC in China is promising because it suggests that China will have at least a marginally improved policy framework for managing the immense challenges it faces. On the economic front, the loss of the demographic dividend threatens to make China old before it gets rich (Chart 7). Xi will need a unified party, as well as loyal supporters in key posts, if he is to re-energize his productivity-enhancing reforms. On the socio-political front, China's intensifying focus on domestic security is symbolized by draconian media censorship ahead of the party congress and, more broadly, a faster rate of spending on public security than national defense in recent years (Chart 8). Such trends suggest that policy makers are concerned about public support. Income inequality and regional disparities are burning issues in an authoritarian country with a larger and more connected middle class and an incipient civil rights movement. Chart 7Rising Participation Boosted Euro Area Labor Force Growth China's Demographic Challenge Rising Participation Boosted Euro Area Labor Force Growth China's Demographic Challenge Rising Participation Boosted Euro Area Labor Force Growth China's Demographic Challenge Chart 8Social Stability A Major Concern In China China's Nineteenth Party Congress: A Primer China's Nineteenth Party Congress: A Primer In terms of the likely economic and market response, we have highlighted in the past that larger macro-economic trends tend to swamp any effects of China's five-year party congresses. There is no observable correlation between these events and the deviations of China's nominal GDP, credit, or fixed investment from long-term averages going back to 1992 (Chart 9). Chart 9No Clear Policy Impact From Past Party Congresses No Clear Policy Impact From Past Party Congresses No Clear Policy Impact From Past Party Congresses Moreover, China only has two midterm party congresses to compare to today's party congress, and both occurred in the thick of global financial crises (1997, 2007). This makes it difficult to draw firm conclusions about any impact on Chinese risk assets. A-shares were mostly flat after the 1997 congress but fell after 2007, while H-shares broadly fell after both meetings, as one might expect given the crises raging around them (Chart 10 A&B). Chart 10AChinese Stocks Were Flat Or Down ... Chinese Stocks Sold Off After Past Midterm Congresses Chinese Stocks Sold Off After Past Midterm Congresses Chart 10B... After Past Midterm Party Congresses Chinese Stocks Sold Off After Past Midterm Congresses Chinese Stocks Sold Off After Past Midterm Congresses H-shares, being highly responsive to global financial market turmoil, fell relative to emerging market (EM) equities as well in 1997 and 2007. A-shares were more insulated and outperformed EM stocks during the 1997 crisis, though not in the 2007 crisis (Chart 11 A&B). What is clear - for Chinese domestic investors - is that A-shares outperformed H-shares after the party congresses in 1997 and 2007 (Chart 12). Chart 11AChinese Stocks Sold Off In Relative Terms... Chinese Stocks Sold Off In Relative Terms... Chinese Stocks Sold Off In Relative Terms... Chart 11B...Except A-Shares During The Asian Crisis ...Except A-Shares During The Asian Crisis ...Except A-Shares During The Asian Crisis Chart 12A-Shares Outperformed H-Shares After Midterm Congresses A-Shares Outperformed H-Shares After Midterm Congresses A-Shares Outperformed H-Shares After Midterm Congresses This fall, it would not be surprising to see Chinese and global risk-on attitudes prevail in the immediate aftermath of China's party congress: in the broadest sense, the meeting represents a political recapitalization for the Xi administration. Moreover, the backdrop is positive: global and Chinese growth are on a synchronized upswing, Chinese industrial profits have improved, the Fed is on hold, and China's growth risks and capital outflow pressures have diminished.16 This suggests a marginal positive impact for H-shares as well as A-shares. However, Chinese stocks are no longer trading at a discount relative to peers. Moreover, BCA's Geopolitical Strategy believes that the Xi administration's reform reboot will likely bring tougher financial and environmental regulation that will slow credit growth and cut into corporate profits.17 It also seems likely that 2018 will see the dollar stage a comeback as inflation recovers and the Fed resumes hiking rates.18 For all these reasons, we recommend staying long Chinese stocks relative to EM, on the basis that China's reform efforts will be positive for China's productivity outlook but negative for commodities and EM in 2018. Matt Gertken, Associate Vice President Geopolitical Strategy mattg@bcaresearch.com 1 Mao's successor Hua Guofeng, and Xi's predecessor Hu Jintao, are the two leaders who did not obtain "core" status. 2 The current norms developed mostly in the 1980s and have evolved since. The list of candidates is mostly pre-arranged by the top leaders. The party congress then votes on which candidates to include, leaving a remainder of about 10% who do not take seats in the Central Committee. 3 Nevertheless, the Central Committee could produce a few surprises. It is almost inevitable that a few major personalities will fail to get promoted into key positions, while others will be catapulted to higher places. There will also be some tea leaves to read about the share of negative votes or abstentions and the implications for different candidates. 4 The political report is filled with arcane Communist Party jargon but is very important. It is a consensus document that takes multiple committees a year or more to draft, though Xi Jinping will give the finishing touches. It will cover a comprehensive range of policies and will be scrutinized closely by experts for slight changes of terminology, emphasis, or omission. Key things to watch for are whether Xi adds or removes entire sections; whether he alters developmental goals outlined in previous administrations; and whether he inserts new concepts or revises party ideology to make way for contentious reforms. As for the party's constitution, the main question of any change is whether Xi's leadership philosophy is incorporated into the Communist Party's guiding thought, and if so, whether Xi's name is explicitly attached to it. The latter in particular would be a sign that Xi's political capital within the party is massive. For additional commentary, please see Alice Miller, "How To Read Xi Jinping's 19th Party Congress Political Report," China Leadership Monitor 53 (2017), available at www.hoover.org. 5 For the "assault stage" of reform, see Robert Lawrence Kuhn, The Man Who Changed China: The Life And Legacy Of Jiang Zemin (NY: Crown, 2004). Jiang had first targeted SOE reform in 1996 in a speech, he launched the policy itself at the party congress in September 1997, and the state began to implement it at the NPC in March 1998. For Hu Jintao's and Wen Jiabao's administrative reforms after the seventeenth party congress, see Willy Wo Lap Lam, "Beijing Unveils Plan For Super Ministries," China Brief, Jamestown Foundation, February 4, 2008. These reforms, which were only part of the overall agenda after the congress, included restructuring the State Council, empowering the National Development and Reform Commission, and setting up "Super-Ministries" to streamline cabinet-level functions. 6 Rumor has it that Xi will keep his anti-corruption chief, Wang Qishan, on the PSC beyond the 69-year mandatory retirement age, and that he could even replace Premier Li Keqiang. We do not expect either to happen, but both are well within the realm of political possibility - particularly retaining Wang. 7 For this estimate, please see Cheng Li, Chinese Politics In The Xi Jinping Era: Reassessing Collective Leadership (Washington, D.C.: Brookings, 2016), chapter 9. 8 Please see BCA Geopolitical Strategy Special Report, "China: Looking Beyond The Party Congress," dated July 19, 2017, available at gps.bcaresearch.com. 9 Traces of Jiang's power will persist here and there, especially if Wang Qishan remains on the PSC, but the overall effect will be a diminishment of this powerful leadership cohort. Symbolically, just as Deng Xiaoping's death loomed over the fifteenth party congress in 1997, Jiang's impending death will loom over the nineteenth party congress today. 10 Indeed judging solely by the cyclical rotation of Chinese leaders according to generation and faction, Hu Jintao's acolytes are favored to outnumber Jiang Zemin's and Xi Jinping's in the 2017 reshuffle. Please see BCA Geopolitical Strategy, "China: Two Factions, One Party," dated September 2012, available at gps.bcaresearch.com. However, Xi's effectiveness and good luck since coming to power lead us to believe that he will secure his followers on the PSC and Politburo this year: please see BCA Geopolitical Strategy Strategic Outlook 2017, "We Are All Geopolitical Strategists Now," dated December 14, 2016, available at gps.bcaresearch.com. 11 For example, Xi Jinping's recent promotions have re-emphasized SOE managers and his policies have supported large "state champion" SOEs. Please see Cheng Li and Lucy Xu, "The rise of state-owned enterprise executives in China's provincial leadership," Brookings, February 22, 2017, available at www.brookings.edu. 12 He came to the top office at a time of significant public dissatisfaction (2012), which meant that he received a kind of "mandate" to make big changes. His faction dominated the PSC, and his sweeping anti-corruption campaign purged the party and state of formidable rivals. In the fall of 2016 he clinched his status as the "core" of the party. 13 As to specific rules, no one should be surprised if they are altered. Take the age limit, which is hotly debated: Jiang Zemin introduced a hard age limit into the PSC in 1997, specifically in a way that prevented the promotion of a heavy-hitting politician, Qiao Shi, while allowing Jiang to continue in power. Now, assume Xi alters the rules to preserve Wang Qishan: this would not necessarily mean that Xi plans to overstay his term limits, though some observers will take it that way. For market participants, the important point is that slight tweaks to informal rules are unlikely to have a big market impact. Consider that Wang has overseen a massive crackdown on corruption, helping clean up the party's image, and is known to be competent in financial regulation as well. If he is retained, will the market really protest? We doubt it. Having said that, we expect him to retire according to the existing rule of thumb. 14 The exception to this statement is if Xi reforms Communist Party political institutions, as some commentators suspect he might, in order to allow the Central Committee to elect the Politburo and PSC directly from its members, thus expanding "intra-party democracy" while also giving Xi a higher likelihood of staying in power. Please see Bo Zhiyue, "Commentary: Sweeping Reforms Expected At Party Congress, But Will Xi Jinping Get All He Wants?" Channel News Asia, August 20, 2017, available at www.channelnewsasia.com. 15 Please see BCA Geopolitical Strategy Monthly Report, "Reflections On China's Reforms," in "The Great Risk Rotation - December 2013," dated December 11, 2013; and Special Report, "Taking Stock Of China's Reforms," dated May 13, 2015, available at gps.bcaresearch.com. Please also see BCA China Investment Strategy, "Understanding China's Master Plan," dated November 20, 2013, available at cis.bcaresearch.com. 16 Please see BCA China Investment Strategy Weekly Reports, "China: Earnings Scorecard And Market Tea Leaves," dated September 7, 2017, and "Monitoring Chinese Capital Outflows And The RMB Internationalization Process," dated August 24, 2017, available at cis.bcaresearch.com. 17 Please see BCA China Investment Strategy Weekly Report, "A Closer Look At Chinese Equity Valuations," dated August 31, 2017, available at cis.bcaresearch.com. For the reform agenda, please see BCA Geopolitical Strategy Special Report, "China: Looking Beyond The Party Congress," dated July 19, 2017, available at gps.bcaresearch.com. 18 Please see BCA Global Investment Strategy Weekly Report, "Central Bank Showdown," dated September 8, 2017, available at gis.bcaresearch.com.

Highlights Bonds As A Safe Haven: Global bond yields have been driven lower by safe haven buying, despite ample evidence of faster global growth and central bankers that are still biased to shift to a less easy policy stance. There is now considerable upside risk for global bond yields over the next 6-12 months from these current depressed levels. ECB: The ECB is giving strong indications that a decision on tapering its asset purchase program will be made next month. With the Euro Area economy growing at a solid pace, and with inflation creeping higher, a reduction in the pace of bond buying in 2018 is highly probable. Canada: The Bank of Canada will continue to deliver on rate hikes without decisive signs that the current booming Canadian economy is slowing down, which leading indicators do not suggest is imminent. Stay underweight Canadian government debt, with a curve flattening bias. Feature Fade The Doomsday Trade Investors have had a lot of depressing news to process over the past several weeks. From threats of nuclear war with North Korea, to fears of a U.S. government shutdown over the debt ceiling, to the potential of Biblical flooding from hurricanes in Texas and Florida, the environment has not been conducive to risk-taking. This has triggered a flight into safe-haven assets like gold and U.S. Treasuries as investors have looked to protect portfolios from "existential" risks (Chart of the Week). Yet despite this rapid run-up in the value of save-havens, risky assets like equities and corporate credit have performed relatively well since the most recent peak in bond yields in early July (Table 1). Chart of the WeekFalling Yields Reflect Save Haven Demand,##BR##Not Slower Growth Falling Yields Reflect Save Haven Demand, Not Slower Growth Falling Yields Reflect Save Haven Demand, Not Slower Growth Table 1Changes In Risk Assets Since##BR##U.S. Treasury Yields Peaked On July 7th Have Bond Yields Peaked For The Cycle? No. Have Bond Yields Peaked For The Cycle? No. This move toward safety and risk aversion has widened the disconnect between global bond yields and economic fundamentals - specifically, growth momentum and central bank guidance - to extreme levels. Investors are now underestimating the potential for additional rate hikes in the U.S. in 2018, and are not fully appreciating the likelihood that the European Central Bank (ECB) will slow the pace of its asset purchases next year. Investors plowing money into government bonds now can only be rewarded if global monetary policy was set to ease, which would only be the case if global growth was slowing. That is not happening right now, even in the U.S. where the most apocalyptic headlines have been occurring. While the impact of Hurricanes Harvey and Irma will likely weigh on U.S. growth in the next few months, the underlying trend remains one of steady above-potential growth that is boosting both corporate profits and household incomes. More globally, depressed investor sentiment, indicated by measures such as the global ZEW survey, has helped drive bond yields lower despite the steady upturn in leading economic indicators (Chart 2). When looking at indicators of actual economic activity, like manufacturing PMIs, the growth story looks far stronger. As a sign of how much this "sentiment versus reality" divergence has distorted bond yields, look no further than our own valuation model for the 10-year U.S. Treasury yield. This model, which only uses the global manufacturing PMI and sentiment towards the U.S. dollar as inputs, indicates that the current "fair value" of the 10-year Treasury yield is 2.67%, nearly 60bps higher than market levels seen as this publication went to press (Chart 3). This is a level of overvaluation that even exceeds the extreme levels seen after the U.K. Brexit vote in July of 2016. Chart 2Bond Investors Are##BR##Ignoring Strong Growth Bond Investors Are Ignoring Strong Growth Bond Investors Are Ignoring Strong Growth Chart 3U.S. Treasuries Are##BR##Now Extremely Overvalued U.S. Treasuries Are Now Extremely Overvalued U.S. Treasuries Are Now Extremely Overvalued In Table 2, we present a decomposition of the 10-year yield changes in the major Developed Markets since that recent peak in U.S. Treasury yields on July 7th. As can be seen in the first two columns of the table, yields declined everywhere but Canada where the central bank has been hiking interest rates (as we discuss later in this report). Yet the vast majority of the yield decline has come from falling real yields and not lower inflation expectations. This has also occurred via a bull-flattening move in government bond yield curves (again, ex-Canada where the curve has bear-flattened), which suggests it is risk-aversion that has driven yields lower. Table 2Developed Market Bond Yield Changes Since U.S. Treasury Yields Peaked On July 7th Have Bond Yields Peaked For The Cycle? No. Have Bond Yields Peaked For The Cycle? No. The relative lack of movement in inflation expectations is a bit surprising given how strongly global oil prices have risen, denominated in any currency (see the final column of Table 2). When plotting the Brent oil price (in local currency terms) vs. the 10-year market-based inflation expectations (from inflation-linked bonds or CPI swaps), some notable divergences stand out. Inflation expectations in the U.S., U.K., Australia and even Japan look around 10-20bps too low relative to where they were the last time oil prices were at current levels (Charts 4 & 5). Meanwhile, inflation expectations are largely in lines with levels implied by oil and currency levels in the Euro Area and Canada. Most importantly, expectations are depressed in all countries, largely because actual inflation has stayed stubbornly low. Chart 4Inflation Expectations Vs. Oil Prices (1) Inflation Expectations Vs Oil Prices (1) Inflation Expectations Vs Oil Prices (1) Chart 5Inflation Expectations Vs. Oil Prices (2) Inflation Expectations Vs Oil Prices (2) Inflation Expectations Vs Oil Prices (2) The lack of realized inflation in places with allegedly "full employment" economies like the U.S. has led to questions over the usefulness of frameworks like the NAIRU (non-accelerating inflation rate of unemployment) in predicting inflation. A reduced link between the NAIRU and inflation does appear in many countries, but not necessarily in all countries when viewed in aggregate. Chart 6The NAIRU Concept Is Not Dead Yet The NAIRU Concept Is Not Dead Yet The NAIRU Concept Is Not Dead Yet In Chart 6, we present an indicator that shows the percentage of OECD economies (34 in total) that have an unemployment rate below the NAIRU rate. Currently, there are 67% of the countries in this list with unemployment rates under the OECD estimate of NAIRU, which is back to levels seen before the 2009 Great Recession. During that pre-crisis period, global inflation rates were accelerating for both goods and services inflation (bottom two panels). While the correlation between this global NAIRU indicator and realized inflation rates declined in the years after the recession, the linkages have improved over the past couple of years. This may be a sign that there is a "global NAIRU level" (or global output gap) that is more important in determining global inflation rates than individual country NAIRU measures. Or put more simply, investors are downplaying the NAIRU concept just at the time when it could be expected to strengthen. If that were the case, inflation expectations around the world would be too low, although it will take some evidence of faster realized inflation (especially in the U.S. and Europe) before the markets begin to discount that in bond yields. In the meantime, markets have become even too pessimistic on growth prospects and the implications for bond yields. Investors have driven down rate hike expectations in the U.S. and U.K. (and, to a lesser extent, the Euro Area) during this latest bond market rally, dragging longer-term bond yields down with them (Chart 7). Yet growth in the developing world is showing little signs of slowing down outside of the U.K., with leading economic indicators still pointing to a continued steady expansion (Chart 8). Even if central bankers are starting to question how fast their economies can grow before inflation pressures pick up in a meaningful way, they are unlikely to stand by and see faster growth prints without responding with less stimulative monetary policies. Chart 7Not Much Tightening Priced##BR##(Except For Canada)... Not Much Tightening Priced (except for Canada)... Not Much Tightening Priced (except for Canada)... Chart 8...Despite Improving Growth##BR##In Most Countries ...Despite Improving Growth In Most Countries ...Despite Improving Growth In Most Countries Net-net, bond markets are now discounting too pessimistic of an outcome for both global growth and inflation. We continue to see more upside risks for global yields on a 6-12 month horizon, although it will take some signs of faster global inflation (not just growth) before bond yields respond. Bottom Line: Global bond yields have been driven lower by safe haven buying, despite ample evidence of faster global growth and central bankers that are still biased to shift to a less easy policy stance. There is now considerable upside risk for global bond yields over the next 6-12 months from these current depressed levels. September ECB Meeting: All Systems Go For A 2018 Taper Last week's ECB meeting provided no changes on interest rates or the size of asset purchases, but plenty of clues on the central bank's next move. A reduction in the size of the ECB's asset purchase program in 2018, to be announced next month, is now highly probable - even with a strengthening euro. The ECB's GDP forecast for 2017 was revised higher from the June forecasts (2.2% vs. 1.9%), while the projections for 2018 (1.8%) and 2019 (1.7%) were unchanged. Meanwhile, the inflation forecast for 2017 was left unchanged at 1.5% and the forecasts for the next two years were only revised slightly lower (2018: 1.2% vs. 1.3%, 2019: 1.5% vs. 1.6%). The fact that the 14% rise in euro versus the U.S. dollar seen so far in 2017 was not enough to move the needle much on the ECB's projections speaks volumes about the central bank's confidence in the current European economic expansion, as well as its comfort level with the rising currency. That makes sense when looking at the euro rally more broadly, as the currency has only gone up 6% in trade-weighted terms year-to-date. Simply put, the ECB does not yet seem overly worried that the strengthening euro represent a serious threat to the economy that could cause a more prolonged medium-term undershoot in Euro Area inflation. ECB President Mario Draghi did make references to currency volatility as being something that should be closely monitored with regards to the growth and inflation outlook. Right now, the realized volatility of the euro has been quite subdued, even as the currency has steadily appreciated (Chart 9). At the same time, our Months-to-Hike indicator has also fallen as the market has pulled forward the date of the next ECB rate hike. That hike is still not expected until late 2019 - pricing that we agree with. However, the fact that the euro can appreciate with such low volatility alongside a slightly-more-hawkish repricing of ECB rate expectations suggests that the market thinks that a move towards reduced monetary stimulus in the Euro Area is credible. That will remain true until the rising euro starts to become a meaningful drag on the economy or inflation, which is not evident in the broad Euro Area data at the moment (Chart 10). Chart 9A "Credibly Hawkish" ECB? A "Credibly Hawkish" ECB? A "Credibly Hawkish" ECB? Chart 10No Impact (Yet) From A Stronger Euro No Impact (Yet) From A Stronger Euro No Impact (Yet) From A Stronger Euro Draghi did note that the "bulk of decisions" regarding the ECB's asset purchase program would likely take place in October. That means a reduction in the size of the monthly purchases starting in January of next year, but without any changes in short-term interest rates (the ECB reiterated that rates will stay at current levels until after the end of the asset purchase program). Nonetheless, the ECB is incrementally moving towards a less accommodative policy stance that will continue to put upward pressure on the euro and, eventually, trigger a move toward higher longer-term Euro Area bond yields. Bottom Line: The ECB is giving strong indications that a decision on tapering its asset purchase program will be made next month. With the Euro Area economy growing at a solid pace, and with inflation creeping higher, a reduction in the pace of bond buying in 2018 is highly probable. Maintain an underweight medium-term stance on Euro Area government debt. Bank Of Canada: Shock Hawks The Bank of Canada (BoC) continues to confound investors with a surprisingly hawkish policy bias. Another 25bp rate hike was delivered at last week's monetary policy meeting, a move that was not fully discounted by the market, bringing the BoC Overnight Rate up to 1%. The Bank cited the impressive strength of the Canadian economy, as well as the more synchronous global expansion that was supporting higher industrial commodity prices, as reasons for the rate hike. With Canadian real GDP growth surging to a 3.7% year-over-year pace in the 2nd quarter, in a broad-based fashion across all components, perhaps policymakers can be forgiven for feeling that interest rate settings are still too stimulative for an economy with a potential growth rate of only 1.4% (the most recent BoC estimate). In the statement announcing the rate hike, it was noted that the level of Canadian GDP was now higher than the BoC had been expecting after the last Monetary Policy Statement (MPS) published in July. The BoC was already projecting that the output gap in Canada would be closed by the end of 2017. Thus, a higher realized level of GDP suggests an output gap that will be closed even sooner than the BoC was forecasting. This alone would be enough to move sooner on rate hikes for a central bank that focuses so much on its own measures of the output gap when making inflation projections. However, at the moment, there is not much inflation for the central bank to worry about. Chart 11The Great White North The Great White North The Great White North Headline CPI inflation sits at 1.2%, well below the midpoint of the BoC's 1-3% target band, while the various measures of core inflation that the BoC monitors are between 1.3% and 1.7%. Annual wage growth accelerated to the faster growth rate of the year in August, but still only sits at 1.7% even with the unemployment rate now down to a nine-year low of 6.2%. Meanwhile, the Canadian dollar has appreciated 13% vs. the U.S. dollar, and 10% on a trade-weighted basis, since bottoming out in early May. This move has been supported by growth and interest rate differentials that favor Canada. This is especially true versus the U.S. where the 2-year gap between Overnight Index Swap (OIS) rates is now positive at +21bps - the highest level since January 2015 (Chart 11). The BoC acknowledged this in last week's policy statement, suggesting acceptance of a strong loonie as a reflection of a robust Canadian economy that requires higher interest rates. The strength in the Canadian dollar will likely weigh on import price inflation in the coming months, and act as a drag on overall inflation. This will not trigger any move by the BoC to back off from its hawkishness unless there is also some weakness in the Canadian economic data. For a central bank that focuses so much on the output gap in its assessment of its own policy stance, the inflationary impact from a booming economy will far outweigh the disinflationary effects of a stronger currency. It remains to be seen if the BoC will be proven right on delivering actual rate hikes with inflation well below target. This is a problem that many central banks are facing at the moment, but the robust Canadian economy is forcing the BoC's hand. An appreciating currency may limit the number of rate hikes that the BoC eventually undertakes, but given its own assessment that that terminal interest rate is around 3%, there are plenty of additional hikes that the BoC can deliver before getting anywhere close to "neutral". The key risk will come from the spillover effects on the overheated Canadian housing market from the interest rate increases. Already, house prices are coming off the boil in the most overheated markets like Toronto, where median home values are down 20% since April due to regulatory changes aimed at reducing leveraged speculation in Canadian housing. It remains to be seen how much the BoC hikes will exacerbate the latest downturn in house price inflation and, potentially, have spillover effects on consumer confidence given high levels of household indebtedness. For now, we do not recommend fighting the BoC, with Canadian leading economic indicators still accelerating and the BoC's own business surveys showing that the economy is likely to remain strong. While there are already 50bps of rate hikes priced next twelve months, this would only take the Overnight Rate to 1.5% - still a stimulative level in the eyes of the central bank. This could also create additional strength in the loonie, although that impact should be lessened if the Fed comes back into play and delivers additional rate hikes in 2018, as we expect. We continue to recommend a below-benchmark duration stance on Canadian government bonds, with yields likely to surpass the relatively modest increases currently priced into the forwards (Chart 12, top panel). We also continue to advise an underweight allocation to Canadian government bonds in hedged global fixed income portfolios (middle panel). We also are staying with our winning Canadian trades in our Tactical Overlay portfolio, where are positioned for wider Canada-U.S. bond spreads and a flatter Canadian yield curve (Chart 13). Chart 12Stay Underweight##BR##Canadian Government Bonds Stay Underweight Canadian Government Bonds Stay Underweight Canadian Government Bonds Chart 13Sticking With Our Tactical##BR##Canadian Bond Trades Sticking With Our Tactical Canadian Bond Trades Sticking With Our Tactical Canadian Bond Trades Bottom Line: The Bank of Canada will continue to deliver on rate hikes without decisive signs that the current booming Canadian economy is slowing down, which leading indicators do not suggest is imminent. Maintain an underweight stance on Canadian government debt, with a curve flattening bias. Robert Robis, Senior Vice President Global Fixed Income Strategy rrobis@bcaresearch.com Have Bond Yields Peaked For The Cycle? No. Have Bond Yields Peaked For The Cycle? No. Recommendations Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Highlights Monetary Policy: A prominent Fed Governor has acknowledged that inflation expectations have become un-anchored to the downside. This is an important signal and suggests that the Fed will keep policy easy enough for inflation expectations to recover. TIPS: The combination of a Fed that communicates a desire for higher inflation expectations and an end to the current downtrend in realized core inflation will send TIPS breakevens wider. Yield Curve: Higher inflation expectations will cause the yield curve to steepen on a 6-12 month horizon. Although steepener trades no longer appear cheap on our model, we remain overweight the 5-year bullet versus a duration-matched 2/10 barbell. Feature Chart 1Flight To Safety Focused In Real Yields Flight To Safety Focused In Real Yields Flight To Safety Focused In Real Yields Bond markets digested two important events last week. The first was a politically driven flight to safety. The 10-year yield fell 10 bps (Chart 1) and the average junk spread widened 8 bps as the daily U.S. Policy Uncertainty index1 averaged 121 for the week, its second-highest reading since February. As we have noted in past reports,2 historically the best strategy has been to fade politically driven flights to safety. The second, and more significant, event was a speech3 given by Fed Governor Lael Brainard in which she suggested that inflation expectations have become un-anchored to the downside. As is explained below, this acknowledgement represents an important change in tone from the Fed. One that reinforces our outlook for higher Treasury yields, a steeper yield curve and wider TIPS breakevens on a 6-12 month horizon. You Had One Job The key passage from Governor Brainard's speech is the following: Nonetheless, a variety of measures suggest underlying trend inflation may be lower than it was before the crisis, contributing to the ongoing shortfall of inflation from our objective. To understand the significance of this statement we need some background on how the Fed thinks about inflation. FOMC members tend to apply an expectations-augmented Phillips curve framework to the task of forecasting inflation (Chart 2). Fed Chair Janet Yellen explained this approach in a September 2015 speech.4 In Yellen's words: ...economic slack, changes in imported goods prices, and idiosyncratic shocks all cause core inflation to deviate from a longer-term trend that is ultimately determined by long-run inflation expectations. [...] An important feature of this model of inflation dynamics is that the overall effect that variations in resource utilization, import prices, and other factors will have on inflation depends crucially on whether these influences also affect long-run inflation expectations. In other words, the Fed's model distinguishes between core inflation's long-run trend and its cyclical fluctuations. Cyclical fluctuations are driven by: Resource utilization (usually measured as the unemployment rate minus its estimated natural rate) Non-oil import prices Idiosyncratic shocks In contrast, core inflation's long-run trend is purely a function of long-term inflation expectations. In the Fed's view, monetary policy can be used effectively in response to shifts in the cyclical drivers of inflation. However, if inflation expectations were to become unanchored, then inflation's long-run trend would be altered and monetary policy would become less effective. In a sense, the worst possible outcome would be if inflation expectations became un-anchored to the downside. Once again, in Janet Yellen's own words: Anchored inflation expectations were not won easily or quickly: Experience suggests that it takes many years of carefully conducted monetary policy to alter what households and firms perceive to be inflation's "normal" behavior, and, furthermore, that a persistent failure to keep inflation under control - by letting it drift either too high or too low for too long - could cause expectations to once again become unmoored. This describes precisely the conventional wisdom as to why the Japanese economy has experienced two decades of deflation despite reasonably high levels of resource utilization. Policymakers did not act quickly or strongly enough following the burst stock market bubble of 1989-91, and this allowed deflationary expectations to become entrenched. In this sense the Japanese experience provides a roadmap for what could happen in the U.S. if the Fed doesn't act quickly to bring inflation expectations back up to target levels. It is true that not all measures of U.S. inflation expectations currently display weakness. For example, the measure we used in our expectations-augmented Phillips curve in Chart 2 - median 10-year PCE expectations from the Survey of Professional Forecasters - appears stable in recent years. However, Governor Brainard pointed to several measures that suggest inflation expectations have already declined (Chart 3). Chart 2The Fed's Inflation Model The Fed's Inflation Model The Fed's Inflation Model Chart 3Still Well Anchored? Still Well Anchored? Still Well Anchored? Comparing the three-year period ending in the second quarter of this year with the three-year period ended just before the financial crisis, 10-year-ahead inflation compensation based on TIPS [...] yields is ¾ percentage point lower. Survey-based measures of inflation expectations are also lower. The Michigan survey measure of median household expectations of inflation over the next five to 10 years suggests a ¼ percentage point downward shift over the most recent three-year period compared with the pre-crisis years, similar to the five-year, five-year forward forecast for the consumer price index from the Survey of Professional Forecasters.5 Investment Implications In our view, there are two important facts to keep in mind: In the Fed's model of inflation it is crucial that long-term inflation expectations do not fall. Otherwise, the odds of replicating the Japanese scenario start to increase. A prominent Fed Governor has now suggested that U.S. inflation expectations have become un-anchored to the downside. Chart 4The Market's Rate Hike Expectations The Market's Rate Hike Expectations The Market's Rate Hike Expectations Taken together, these two facts have important investment implications. First, the two facts suggest that TIPS breakevens will move wider. While the Japanese experience has taught us that "open mouth operations" become less effective once deflationary expectations are entrenched, they should still have some impact in the States. Notice that the decline in Treasury yields that followed Brainard's comments last week was concentrated in the real component. The 10-year TIPS breakeven inflation rate actually rose 2 bps (Chart 1). The combination of a Fed that communicates a desire for higher inflation expectations and an end to the current downtrend in realized core inflation (see "Economy & Inflation" section below) will be enough to send long-dated TIPS breakevens wider on a 6-12 month horizon. Second, a Fed that is committed to staying accommodative for as long as is necessary to ensure that inflation expectations move higher will cause the yield curve to steepen (see section titled "Inflation Expectations Drive The Curve" below). Third, a Fed that is more committed to fighting deflation should bias Treasury yields lower. However, inflationary pressures in the U.S. economy are strong enough that the Fed will be able to move inflation expectations higher while still delivering more rate hikes than are currently priced into the curve. At present, the overnight index swap curve is discounting that the next 25 basis point rate hike will not occur until November 2018 (Chart 4)! Bottom Line: A prominent Fed Governor has acknowledged that inflation expectations have become un-anchored to the downside. This represents an important signal about the future path of policy and reinforces our view that the Treasury curve will bear-steepen during the next 6-12 months, led by wider TIPS breakevens. Inflation Expectations Drive The Curve Our research6 shows that inflation expectations are the most important driver of changes in the slope of the yield curve. This runs counter to the conventional wisdom which states that the curve flattens when the Fed hikes rates, and steepens when it cuts rates. While the correlation between Fed rate moves and the slope of the curve is undeniable, the relationship results purely from the fact that the Fed responds to changes in inflation. The link between inflation expectations and the yield curve is the dominant relationship. To see this we look at Charts 5 and 6. Both charts show monthly changes in the 5-year, 5-year forward TIPS breakeven inflation rate plotted against monthly changes in the nominal 2/10 slope. Chart 5 shows all available historical data, and we observe a strong positive correlation. In fact, 63% of monthly observations fall into either the top-right or bottom-left quadrants indicating that wider breakevens correlate with a steeper curve and vice-versa. Chart 52/10 Nominal Treasury Slope Vs. TIPS Breakeven Inflation Rate 5-Year / ##br##5-Year Forward (February 1999-Present) Open Mouth Operations Open Mouth Operations Chart 62/10 Nominal Treasury Slope Vs. TIPS Breakeven Inflation Rate 5-Year / 5-Year Forward ##br##During Fed Tightening Cycles (June 1999 To May 2000 & June 2004 To June 2006) Open Mouth Operations Open Mouth Operations The more important question, however, is whether this correlation still holds when the Fed is raising rates. Chart 6 focuses only on prior rate hike cycles and still shows a strong positive correlation. 73% of the monthly observations fall into either the top-right or bottom-left quadrants, although in this case there are more observations in the bottom-left quadrant because typically the Fed lifts rates with the goal of sending inflation and inflation expectations lower. In this respect the current rate hike cycle is unique. The Fed is in the process of lifting rates, but as Brainard's speech shows, it still critically needs inflation expectations to rise. We conclude that the Fed will stay easy enough, long enough, for long-dated TIPS breakevens to return to their pre-crisis trading range between 2.4% and 2.5%. An upward adjustment to this range will occur alongside a steeper 2/10 curve. Unit Labor Costs And The Yield Curve The logic presented above also suggests an inverse relationship between the slope of the curve and wage growth. In a world where inflation expectations are well anchored, stronger wage growth encourages the Fed to tighten policy more quickly, this causes the yield curve to flatten. Conversely, softer wage growth leads to a steeper curve. Our research shows that unit labor costs are the measure of wage growth that correlates most closely with the slope of the curve. The reason is that unit labor costs actually measure both wage growth (compensation per hour) and labor productivity (output per hour). Put differently, the yield curve can flatten because labor compensation is rising and the Fed is tightening policy (bear flattening) or it can flatten because productivity is falling and investors are discounting a slower pace of potential growth and a lower terminal fed funds rate (bull flattening). Unit labor costs capture both of these dynamics. Last week saw second quarter productivity growth revised higher from 0.9% to 1.5% and unit labor cost growth revised down from 0.6% to 0.2% (Chart 7). We expect that productivity will continue to experience a modest late-cycle bounce. Usually, payroll growth starts to moderate late in the business cycle as the labor market tightens. The cost of labor typically rises and encourages firms to substitute capital for workers. This late-cycle boost in capital spending tends to correlate with stronger productivity growth (Chart 8), and this dynamic looks to be in full swing at the moment. Payroll growth has been decelerating since early 2015, and durable goods orders have picked up sharply since the end of last year (Chart 8, bottom panel). Chart 7Weakness In Unit Labor Costs Weakness In Unit Labor Costs Weakness In Unit Labor Costs Chart 8Productivity: Look For A Late-Cycle Rebound Productivity: Look For A Late-Cycle Rebound Productivity: Look For A Late-Cycle Rebound A modest late-cycle upswing in productivity growth will put downward pressure on unit labor costs and lead to curve steepening. How To Position For Steepening We have been expressing our yield curve view via a long position in the 5-year bullet and a short position in a duration-matched 2/10 barbell since last December.7 So far that trade has returned +28 bps, even though the 2/10 slope has flattened more than 50 bps since its inception. The reason our curve steepener has outperformed even as the curve has flattened is that, when we initiated our trade, the 2/5/10 butterfly spread was discounting an even larger curve flattening. Put differently, the 5-year bullet looked extremely cheap on the curve (Chart 9).8 Chart 92/5/10 Butterfly Spread Fair Value Model 2/5/10 Butterfly Spread Fair Value Model 2/5/10 Butterfly Spread Fair Value Model This state of affairs has now changed. Our fair value model shows that the 5-year bullet appears slightly expensive compared to the barbell, or alternatively, that the 2/5/10 butterfly spread is priced for a 20 bps steepening of the 2/10 slope during the next six months. According to our model, the 2/10 slope will have to steepen by more than 20 bps during the next six months for our trade to outperform from current levels. Bottom Line: Higher inflation expectations will cause the yield curve to steepen on a 6-12 month horizon. Although steepener trades no longer appear cheap on our model, we remain overweight the 5-year bullet versus a duration-matched 2/10 barbell for now. Economy & Inflation Updates received during the past few weeks indicate that U.S. growth is running solidly above trend, and may even be accelerating. Real second-quarter GDP growth was revised higher from 2.6% to 3%. Second quarter labor productivity growth was also revised higher, as was discussed above. Even following a lackluster August employment report, our back-of-the-envelope tracking estimate for U.S. growth - the sum of year-over-year growth in aggregate hours worked and average quarterly productivity growth since 2012 - is running at 2.7%, well above the Fed's 1.8% estimate of trend (Chart 10). Survey measures also suggest that growth has further upside in the second half of the year, at least according to a simple growth model based on the ISM non-manufacturing survey, our own BCA Beige Book Monitor and a composite of new orders surveys (Chart 11). Chart 10Growth Tracking Above-Trend... Growth Tracking Above-Trend... Growth Tracking Above-Trend... Chart 11...And Surveys Suggest Further Upside ...And Surveys Suggest Further Upside ...And Surveys Suggest Further Upside But bond markets are not getting the message. The 10-year yield is stuck at 2.12%, and the markets seem to be saying that the link between stronger growth and rising inflation has been permanently broken. We disagree and think that investors are simply underestimating the often long and variable lags between economic growth and inflation. Chart 12Inflation Lags Growth Inflation Lags Growth Inflation Lags Growth Chart 12 shows that real GDP growth has tended to lead core inflation by about 18 months, while changes in year-over-year core CPI (the second derivative of prices) have tended to follow the ISM Manufacturing index with a lag of about 12 months. All signs suggest that the recent downtrend in inflation is nothing more than a reaction to the growth deceleration seen between mid-2015 and mid-2016. Now that growth has re-accelerated, inflation is poised to move higher. Bottom Line: Bond markets are priced as though the link between growth and inflation is broken. We expect they will be proven wrong as inflation regains its uptrend during the next few months. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 The daily policy uncertainty index measures the number of news items related to economic uncertainty. For further details please see www.policyuncertainty.com 2 Please see U.S. Bond Strategy Weekly Report, "What We Know About Uncertainty", dated July 12, 2016, available at usbs.bcaresearch.com 3 https://www.federalreserve.gov/newsevents/speech/brainard20170905a.htm 4 https://www.federalreserve.gov/newsevents/speech/yellen20150924a.htm 5 https://www.federalreserve.gov/newsevents/speech/brainard20170905a.htm 6 Please see U.S. Bond Strategy Weekly Report, "The Yield Curve On A Cyclical Horizon", dated March 21, 2017, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy Special Report, "Seven Fixed Income Themes For 2017", dated December 20, 2016, available at usbs.bcaresearch.com 8 For further details on how butterfly trades respond to changes in the yield curve, and on how we use our fair value yield curve models please see U.S. Bond Strategy Special Report, "Bullets, Barbells And Butterflies", dated July 25, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Portfolio Strategy A supply/demand imbalance has created a playable opportunity in the niche refining energy sub-index. Increase exposure to overweight. Safe haven demand is supporting gold mining equities, but shifting macro forces suggest that it will soon be time to move to the sidelines. Global gold miners are now on downgrade alert. Recent Changes Lift the S&P oil & gas refining & marketing index to overweight today. Put the global gold mining equity index (ticker GDX:US) on downgrade alert. Table 1 Still Goldilocks Still Goldilocks Feature The S&P 500 moved laterally last week as sustained geopolitical uncertainty offset encouraging economic data. Synchronized global growth coupled with the related global liquidity-to-growth transition remain the dominant macro themes. Dovish Fed speeches triggered a recalibration of market rate hike expectations and a lower 10-year Treasury yield. As long as lower bond yields reflect a less hawkish Fed rather than a deflationary relapse, they should underpin stock prices. Encouragingly, the latest ISM manufacturing survey catapulted higher to a level last seen in early 2011, diverging steeply from the bond market, as manufacturing optimism reigns supreme (Chart 1). The labor market confirmed this data. The most cyclical parts of the U.S. economy are firing on all cylinders, with manufacturing and construction job creation comprising 1/3 of nonfarm payroll growth last month (Chart 2). This is the highest reading since July 2011. Chart 1Unsustainable Divergence Unsustainable Divergence Unsustainable Divergence Chart 2Manufacturing Flexing Its Muscle Manufacturing Flexing Its Muscle Manufacturing Flexing Its Muscle Meanwhile, despite the Trump administration's shortcomings, America's CEOs are going against the grain. Capex is up smartly for the second consecutive quarter adding to real GDP growth and our capital spending model remains upbeat heralding additional outlays for the remaining two quarters of the year (Chart 3). Similarly, regional Fed surveys of capex intentions point to a sustainable pickup in capital spending in the coming months (Chart 3). Still generationally low interest rates, a less hawkish sounding Fed, coupled with a tamed greenback (Chart 4) and synchronized global growth have combined to revive animal spirits. The implication is that profit growth rests on solid foundations, a message corroborated by our S&P 500 EPS growth model (Chart 5). Chart 3CapEx To The Rescue CapEx To The Rescue CapEx To The Rescue Chart 4Dollar... Dollar… Dollar… Chart 5...And EPS Model Waving Green Flag …And EPS Model Waving Green Flag …And EPS Model Waving Green Flag Adding it up, the macro backdrop remains favorable for stocks. In fact, it represents a goldilocks equity scenario. This week we continue to add some cyclicality to our portfolio by further boosting a niche energy play. We also update our view on a portfolio hedge. Buy Refiners For A Trade In early July, we lifted refiners to neutral and locked in impressive gains for our portfolio, but three reasons kept us at bay and prevented us from turning outright bullish on this niche energy sub-sector.1 Namely, all-time high refining production, high refined product stocks and breakneck pace refinery runs were offsetting the nascent recovery in gasoline consumption, rising crack spreads and a mini V-shaped recovery in industry shipments. Net, we posited that a balanced EPS outlook would prevail in coming quarters. Hurricane Harvey has significantly changed this calculus and now clearly refiners are in a sweet earnings spot for at least the remainder of the year, compelling us to lift exposure to overweight. Severe refinery shutdowns are likely to return industry production levels to what prevailed early in the decade, representing a major, albeit temporary, setback (Chart 6). This production curtailment will result in sizable petroleum products inventory drawdowns and a likely halt (if not reversal) in refined product net exports in order to satisfy domestic demand. The longer it takes for refinery production to return to normalcy, the greater the inventory whittling down. Historically, relative share price momentum has been inversely correlated with inventory growth and the Harvey-related inventory clear-out is heralding additional relative performance gains (bottom panel, Chart 7). It is notable that both industry net exports and inventories had already been receding since the beginning of 2017, suggesting that hurricane Harvey will only accelerate a downtrend that was already in place. Chart 6Hurricane Related Blues... Hurricane Related Blues… Hurricane Related Blues… Chart 7... Are A Boon For Crack Spreads … Are A Boon For Crack Spreads … Are A Boon For Crack Spreads Taken together, this represents an ultra-bullish pricing power backdrop for the U.S. refining industry, at a time when capacity additions are also likely to, at least, pause for breath (bottom panel, Chart 6). Chart 8Brisk Demand Brisk Demand Brisk Demand Indeed, refining margins have jumped recently and will likely remain elevated as the Brent/WTI spread is widening anew (middle panel, Chart 7). Surging crack spreads are synonymous with higher earnings for this extremely capital-intensive and high operating leverage industry. Nevertheless, the refining supply disruptions only tell half the story. Refined product demand is exploding higher, pushing all-time highs and signaling that a substantial supply/demand imbalance is in the works (top panel, Chart 8). Typically this gets resolved via higher gasoline prices, further boosting industry EPS prospects (third panel, Chart 8). As a result, we expect a re-rating phase in relative valuations in the coming months, reversing the year-to-date deflation in the relative price-to-sales ratio. The second panel of Chart 8 shows that relative valuations and refined product consumption move in lockstep, and the current message is to expect a catch up phase in the former. In sum, a playable rally in refiners is in the offing on the back of a budding profit recovery that has yet to filter through analysts' EPS estimates (bottom panel, Chart 8). The longer-than-usual hurricane Harvey-related refining production disruptions, along with the spike in refined product demand, have created an exploitable opportunity. Bottom Line: Boost the S&P oil & gas refining & marketing index (PSX, VLO, MPC, ANDV) to overweight. What To Do With Gold Mining Equities? Gold and gold mining equities serve as great portfolio hedges especially in times of duress. Recent geopolitical jitters surrounding North Korea along with inaction in Washington and the substantial year-to-date selloff in the U.S. dollar have served as catalysts for gold to shine anew, hitting one-year highs. So is it time to trim exposure to shiny metal equities? The short answer is not yet. Real yields are sinking courtesy of a moderately less hawkish Fed (top panel, Chart 9). The probability of a December Fed hike has now collapsed to 30%, and the 5th hike this cycle is only priced in for next June. This is keeping a bid under gold and gold miners, as zero yielding bullion and near-zero yielding gold mining equities appear at the margin relatively more appealing. The equity risk premium has also stopped falling owing largely to the lower 10-year Treasury yield (bottom panel, Chart 9), representing another source of support for global gold miners. Meanwhile, policy uncertainty in the U.S. and around the globe is hooking up especially given North Korea's unpredictability, Washington's polarization, the upcoming German elections and, most importantly, the looming Chinese Congress. Historically, the policy uncertainty index and relative performance have been joined at the hip and the current message is positive for bullion related stocks (middle panel, Chart 9). Similarly, the Philly Fed's Partisan Conflict Index2 ("The Partisan Conflict Index tracks the degree of political disagreement among U.S. politicians at the federal level by measuring the frequency of newspaper articles reporting disagreement in a given month. Higher index values indicate greater conflict among political parties, Congress, and the President.") and bullion enjoy a tight positive correlation since the early 1980s (Chart 10), likely warning that the precious metal's run has more upside in the short term. Chart 9Shining Shining Shining Chart 10Increase In Partisanship Is Bullish Gold Increase In Partisanship Is Bullish Gold Increase In Partisanship Is Bullish Gold Moreover, demand for safe haven assets remains upbeat as evidenced by recent flows into gold-related ETFs. Positioning in the commodity pits are also signaling that more gains are in store for gold and the relative share price ratio (Chart 11). Nevertheless, there are some pockets of weakness that are pointing to a more cautious stance toward this portfolio hedge. The improving U.S. economic backdrop is weighing on gold mining equities (ISM manufacturing shown inverted, middle panel, Chart 12). Not only U.S. growth, but also synchronized global growth suggests that eventually demand for bullion will subside. In fact, global growth expectations continue to perk up (GDP expectations shown inverted, Chart 12), and G10 economic surprises are also shooting higher, anchoring gold and gold related equities (economic surprise index shown inverted, top panel, Chart 12). Chart 11Safe Haven Demand Comeback Safe Haven Demand Comeback Safe Haven Demand Comeback Chart 12Not All The Glitters Is Gold Not All The Glitters Is Gold Not All The Glitters Is Gold Tack on the inevitable liquidity withdrawal once the Fed starts to wind down its balance sheet later this month, and the handoff from liquidity-to-growth represents a bearish backdrop for gold and gold mining equities. Chart 13 shows that the Fed's balance sheet is positively correlated with bullion's relative performance versus the broad commodity complex, warning that the recent push toward multi-decade highs in relative performance are on borrowed time. Finally, our relative EPS model for the global gold mining index encapsulates most of these macro forces and suggests that relative profit growth will gravitate lower in the coming months (Chart 14). Chart 13Watch The Fed's Balance Sheet Watch The Fed’s Balance Sheet Watch The Fed’s Balance Sheet Chart 14EPS Model Is Outright Bearish EPS Model Is Outright Bearish EPS Model Is Outright Bearish Bottom Line: While our confidence in maintaining the gold-related equity portfolio hedge has fallen a notch, we are staying patient before moving to the sidelines. Put the global gold mining index (ticker GDX:US) on downgrade alert. Anastasios Avgeriou, Vice President U.S. Equity Strategy & Global Alpha Sector Strategy anastasios@bcaresearch.com 1 Please see the July 10, 2017 U.S. Equity Strategy Report titled "SPX 3,000?", available at uses.bcaresearch.com 2 https://www.philadelphiafed.org/research-and-data/real-time-center/partisan-conflict-index Current Recommendations Current Trades Size And Style Views Favor small over large caps and stay neutral growth over value.

Dear Client, I have been visiting clients in Europe this week, so today's report is somewhat shorter than usual. We will be back next week with an exciting Special Report on the macro effects of bitcoin and other virtual currencies. Best regards, Peter Berezin, Chief Global Strategist Global Investment Strategy Highlights Global growth remains strong and broad-based. U.S. GDP growth will accelerate over the next few quarters thanks to the easing in financial conditions so far this year. The market is pricing in only 20 basis points in cumulative rate hikes between now and the end of next year. This is far too low. Go short the Dec-2018 fed funds futures contract. The euro has strengthened more this year than one would have expected based solely on the change in interest rate differentials. Positioning shifts are the likely culprit. In real terms, the terminal rate in the U.S. based on 5-year, 5-year forward OIS rates is currently only 13 bps higher than in the euro area. We will automatically open a tactical short EUR/USD position if the euro moves above $1.22 any time over the next three weeks. Feature Global Economy Firing On All Cylinders The global economy continues to chug along. All 46 countries monitored by the OECD are on track to grow this year, the first time this has happened since 2007. Usually, economists are too optimistic about growth prospects. This has not been the case over the past 12 months. Consensus global growth estimates for 2017 and 2018 have marched higher during this time, led by the euro area, Japan, and a number of emerging economies (Chart 1). U.S. growth projections have been broadly stable, but these too are likely to be revised higher. Both the manufacturing and non-manufacturing ISM indices improved in August. The same goes for core capital goods orders, consumer confidence, retail sales, and homebuilder sentiment. The employment report was on the weak side, but it was probably distorted by seasonal factors - August payrolls have now fallen short of expectations for seven years in a row, a suspiciously long streak. Hiring intention surveys and perceptions of job availability both remain strong. The net share of households who see jobs as "plentiful" as opposed to "hard to get" rose further in August. It is now well above its pre-recession peak (Chart 2). Chart 1Higher And Higher Higher And Higher Higher And Higher Chart 2A Healthy U.S. Labor Market A Healthy U.S. Labor Market A Healthy U.S. Labor Market A Healthy U.S. Labor Market A Healthy U.S. Labor Market Fed Rate Expectations Are Too Dovish The Treasury market remains oblivious to these developments, focusing only on the failure of inflation to rise. This could prove to be a fatal mistake. Inflation is a highly lagging indicator. It typically does not peak until well after a recession has begun and does not bottom until well after it has ended (Chart 3). Trying to infer the true level of economic slack from today's inflation rate is like trying to read the speedometer of an automobile when there is a 30-second delay between what the dial says and when you step on the accelerator. Many market participants and a number of Fed officials have argued that interest rates are already close to neutral, implying little need for further rate hikes. We agree that the neutral rate is lower than in the past, but their argument misses a crucial point. Even if the Fed knew what the level of the neutral rate is - which, of course, it doesn't - it would still need to get the timing right. If the Fed waits too long to bring rates up to neutral, the unemployment rate will end up falling below NAIRU. This could force the Fed to raise rates more aggressively than it (or the markets) would like. Such an outcome now looks increasingly likely. The easing in U.S. financial conditions since the start of the year should boost real GDP growth over the next few quarters (Chart 4). This could cause the unemployment rate to fall to 3.5% by next summer, leaving it below its 2000 lows and more than a full point below most estimates of NAIRU. If this were to happen, it would prompt the Fed to turn up the hawkish rhetoric. Chart 3Inflation Is A Lagging Indicator Central Bank Showdown Central Bank Showdown Chart 4Easing Financial Conditions In The U.S. Bode Well For Growth Easing Financial Conditions In The U.S. Bode Well For Growth Easing Financial Conditions In The U.S. Bode Well For Growth The market is not giving enough weight to such an outcome. The December-2018 fed funds futures contract is pricing in only 20 basis points in cumulative rate hikes between now and the end of next year. That is much too low. We recommend that clients short this contract and are initiating a new tactical trade to this effect. ECB Will Take It Easy In contrast to the U.S., euro area financial conditions have tightened this year. During his press conference, Mario Draghi expressed confidence in the growth outlook, but acknowledged the risks to the region from a stronger currency. He noted that "the recent volatility in the exchange rate represents a source of uncertainty which requires monitoring with regard to its possible implications for the medium-term outlook for price stability." As we predicted last week, the ECB trimmed its 2018 inflation forecast from 1.3% to 1.2%, and its 2019 forecast from 1.6% to 1.5%. Chart 5 shows the market's estimate of the gap in terminal interest rates between the U.S. and the euro area using 5-year, 5-year forward OIS rates. The gap has narrowed by around 50 bps since the start of the year. However, EUR/USD has strengthened more than one would have expected based solely on the movement in interest rate differentials. Specifically, the market now expects U.S. five-year yields to be 78 basis points higher in 2022 than in the euro area. This is precisely the same gap that prevailed last October. Yet, EUR/USD was $1.10 back then. Today, it is $1.20. Shifts in positioning help explain why the euro has strengthened so much. Traders were heavily short the euro at the start of this year. Today, they are heavily long (Chart 6). Looking out, with few euro shorts left, EUR/USD is likely to trade off the interest rate gap between the two regions. Chart 5U.S. Vs. Euro Area: Interest Rate Gap Has Narrowed U.S. Vs. Euro Area: Interest Rate Gap Has Narrowed U.S. Vs. Euro Area: Interest Rate Gap Has Narrowed Chart 6Euro Positioning: From Deeply Short To Long Euro Positioning: From Deeply Short To Long Euro Positioning: From Deeply Short To Long Chart 7Fiscal Policy Is More Stimulative In The U.S. Central Bank Showdown Central Bank Showdown In real terms, the terminal rate in the U.S. is currently only 13 bps higher than in the euro area. That seems rather low to us. Trend growth is faster in the U.S., the banking system is in better shape, and fiscal policy is more stimulative (Chart 7). All this suggests that the real neutral rate is substantially higher in the U.S. As such, we will automatically open a tactical short EUR/USD position if the euro moves above $1.22 at any time over the next three weeks, with a stop of $1.24 and a year-end target of $1.15. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights Based on long-term moving averages and the advance/decline line, the dollar selloff is still only a severe correction. These factors need to be monitored closely as they stand on the edge. To rebound, the dollar will need U.S. inflation to pick up, which will lift the U.S. OIS curve. Signs are accumulating that U.S. inflation will trough toward the end of 2017. Buying the dollar versus the yen is a much safer bet than shorting the euro. The CAD has more upside, especially on its crosses. Feature The U.S. dollar continues to be tested by investors. As paradoxical as it may sound, it is still too early to sound the death knell for the dollar bull market. However, it is not time either to aggressively bet on a rebound. For that to happen, U.S. inflation must regain its footing in a more convincing fashion. Why Isn't The Bull Market Dead? There are many facets to this question, but let's begin with technical considerations. First, the dollar's advance / decline line has not broken down (Chart I-1). A breakdown in this measure would be one of the key technical signals that the dollar has begun a new cyclical downtrend. In the mid-1990s, the dollar did experience a period of correction. During that time frame, the A/D line was also unable to break down, later highlighting that what was initially perceived as the beginning a new bear market was ultimately a prolonged period of consolidation. Chart I-1Still Not A Cyclical Bear Market Still Not A Cyclical Bear Market Still Not A Cyclical Bear Market Second, the dollar's trend has been best approximated by the four-year moving average of monthly prices. Since the Smithsonian Agreement of 1971, during bear markets, the dollar tends to find its ceiling around this indicator, and during bull markets, it tends to put a floor around this moving average (Chart I-2). Today, the dollar has yet to end a month below this measure. Third, positioning in the dollar is now depressed, as investors have purged their stale USD longs (Chart I-3). When one looks at net long speculative positions in EUR/USD - the most convenient and liquid instrument to bet on the dollar - investors are clearly enamored with the euro, which by definition illustrates their dislike of the greenback. Chart I-2No Trend Break For Now No Trend Break For Now No Trend Break For Now Chart I-3Dollar Downside Is Limited Dollar Downside Is Limited Dollar Downside Is Limited Technical indicators argue that we have experienced a painful correction in the USD, but valuation considerations suggest it will be difficult for these technical indicators to deteriorate enough to begin flagging a cyclical bear market. Our long-term fair value model, which incorporates productivity differentials, highlights that the dollar never hit the nosebleed levels associated with bull market tops in 1985 or in 2001 (Chart I-4). The stability in the trade balance and the current account - both have been stable at around 3% of GDP and 2.5% of GDP, respectively - are at odds with the sharp deterioration in the balance of payments that has occurred when the dollar has been genuinely expensive. Our intermediate-term valuation models point to an even more unequivocal conclusion. Based on this metric, the DXY is at its cheapest level since 2009, a discount that historically has been associated with dollar bottoms, at least temporary ones (Chart I-5). This gives us comfort that the A/D line is unlikely to break down for now, or that the dollar will end September significantly below its crucial four-year moving average. However, if these things happen, the dollar could experience significant downside. Chart I-4The Dollar Never Reached Nosebleed Valuations The Dollar Never Reached Nosebleed Valuations The Dollar Never Reached Nosebleed Valuations Chart I-5Big Discount To IRP Big Discount To IRP Big Discount To IRP Economic forces too do not point to a sharp move in the DXY below 91 - one that could drive the dollar down into the low 80s. After a period of deep underperformance, the U.S.'s economic surprises relative to the G10 have begun to stabilize, as have inflation surprises. More saliently, the incredible strength in the U.S. ISM manufacturing index, especially when compared to other PMIs around the world, points to a rebound in the USD, or at the very least, stabilization (Chart I-6). Finally, the market has now all but priced out additional hikes from the U.S. interest rate curve. There are only 30 basis points of hikes priced in over the next 24 months. Moreover, the probability of the fed funds rate remaining between 1% and 1.25% only falls below 50% in September 2018 (Table I-1). This seems to be a sanguine scenario. Chart I-6Cyclical Support ##br## For USD Cyclical Support For USD Cyclical Support For USD Table I-1Investors See U.S. Rates At Current ##br##Levels Until Late 2018 Conflicting Forces For The Dollar Conflicting Forces For The Dollar Bottom Line: The dollar's technicals are not yet indicative of the end of the cyclical bull market. However, they do need to be monitored closely. Additionally, the dollar is trading at a large discount to interest rate parity relationships, and the Federal Reserve is not expected to execute its next hike until late 2018. While these factors may not point to an imminent rebound in the USD, they do suggest that the down-wave in the dollar is very long in the tooth. Chasing the dollar lower is dangerous. Too Early To Bet The House On A Renewed Upleg Chart I-7The Global Deflation Anchor The Global Deflation Anchor The Global Deflation Anchor This observation on the probability of a Fed move brings us to the vital question of what could lift the U.S. interest rate curve higher, and thus the dollar. This would be the outlook for inflation. As Fed Governor Lael Brainard clearly argued this week, the Fed is not meeting its inflation mandate, warranting a slower pace of rate increases as global deflationary forces remain very potent. The dovish path implied by interest rate markets shows that investors already agree with this assessment. There is no denying that inflation has been globally and structurally pulled down by various forces. While the "Amazon effect" has grabbed headlines, Mark McClellan argues in The Bank Credit Analyst this month that the effect of e-commerce on inflation is no greater than that of Walmart in the 1990s - and probably amounts to a meagre 0.1-0.2% depressive impact on inflation.1 Instead, we peg the capacity buildup in EM and China - which has lifted the global capital stock massively since the turn of the millennia - as the main source of global deflation (Chart I-7). Now that global credit growth is lower than it was before 2008, it has become clearer that the global supply side of the economy has expanded faster than underlying demand, resulting in downward pressure on prices. Nonetheless, while there is a lid on inflation, this does not imply that cyclical determinants of inflation have been fully neutered. They simply have become weaker. Inflation can still ebb and flow in response to the business cycle, but the upside is not as strong as it once was. This limits how high nominal interest rate can go, which is why it is hard to envision a terminal rate much above 3% - a very low reading by post-war standards. Here, we continue to see a turning point coming later this year for inflation, one that would pull core PCE closer to the 2% mark wanted by the Fed in 2018. In the background, our composite capacity utilization indicator is now firmly in "no slack" territory, an environment in which inflation tends to perk up and where interest rate exhibit upside (Chart I-8). This is not enough to warrant fears of inflation, but healthy growth in this context should be a red flag for deflationists. This is exactly the set of circumstances we envision for the next 12 months, even if hurricane Harvey and its potential successors create noise in upcoming data. The U.S. economy has benefited from a strong easing in financial conditions since February 2016. The recent fall in real rates, which has been the key driver of the 60 basis-points fall in Treasury yields since December 2016, is now demonstrably reflationary. Lumber prices are once again at the top of their trading range since 2013, and gold prices have regained vigor. In this optic, the ratio of metal to bond prices - adjusted for their very different volatilities - has been a reliable leading indicator of U.S. growth (Chart I-9). Today, it is pointing to an acceleration of GDP growth relative to potential, the very definition of declining slack. Chart I-8Tight U.S. Capacity Is Inflationary Tight U.S. Capacity Is Inflationary Tight U.S. Capacity Is Inflationary Chart I-9Relfation Will Boost U.S. Growth Above Trend Relfation Will Boost U.S. Growth Above Trend Relfation Will Boost U.S. Growth Above Trend The labor market continues to display signs of resilience as well. True, the last employment report was paltry, but August has been marked by seasonal weaknesses for the past seven years. Moreover, August weaknesses have tended to be minimized in the wake of the notorious revisions typical of the U.S. Department of Labor. However, the strength in the labor market components of the NFIB small businesses survey highlights the potential for more job gains going forward. Where this indicator really shines though, is in its capacity to forecast household total wages and salaries (Chart I-10). Today, this gauge highlights that the income of middle class households will accelerate over the next six months. This matters because if the middle class - a category of U.S. households that gather the vast majority of their income from wages - experiences strong income growth, this will create robust support for consumption. With consumption accounting for 70% of U.S. GDP, a boost to this component would go a long way in lifting aggregate growth. Stronger growth in a tight economy is inflationary, and monetary dynamics confirm this risk. The U.S. velocity of money has picked up meaningfully, and now suggests that inflation will gather steam later this year (Chart I-11). Chart I-10The Labor Market Is Still Strong The Labor Market Is Still Strong The Labor Market Is Still Strong Chart I-11Monetary Dynamics Point To More Inflation Monetary Dynamics Point To More Inflation Monetary Dynamics Point To More Inflation We therefore expect that when this turnaround in inflation emerges, investors will re-assess their expectations for the path of U.S. monetary policy, and the dollar will finally be able to resume its upward trek toward new highs. But until inflation turns the corner, the dollar will continue to struggle to rally durably. Bottom Line: The U.S. economy is still on a firming path. With the amount of slack in the economy vanishing and with the velocity of money accelerating, this will lead to a pick-up in inflation late this year. The end of Q4 is likely to prove the moment when the dollar will finally be able to begin firming up. Investment Implications Shorting the Yen Is Still The Safest Bet Shorting the yen remains the best way to play a dollar rebound for now. The yen has not benefited much from the recent bout of risk aversion prompted by the renewed flare-up of in tensions in the Korean peninsula. It remains weak on its crosses like EUR/JPY, CAD/JPY or even AUD/JPY. USD/JPY seems incapable of staying below 108.5, and may even be forming a consolidation pattern reminiscent of the one experienced in 2013 (Chart I-12). In late 2013, this pattern was resolved by U.S. bond yields moving higher. This time is likely to be similar. The recent weakness in Japanese wages remains a key hurdle that the Bank of Japan does not seem able to shake off. Wage growth hit it slowest pace since 2015 and real wages are worryingly weak (Chart I-13). This is not the picture of an economy with any hint of inflation, even if the labor market is tight. Illustrating this point, contrarily to the euro area, Japanese inflation expectations have not kept pace with the U.S. in recent months (Chart I-14). This argues that the BoJ faces the greatest burden of any central bank. With the BoJ now packed with doves, we expect that interest rates and bond yields in Japan will remain capped for the foreseeable future. As a result, if U.S. bond yields can rise in the face of a strong U.S. economy, JGB yields will not follow higher. This will flatter USD/JPY. Chart I-12Consolidation Pattern In USD/JPY Consolidation Pattern In USD/JPY Consolidation Pattern In USD/JPY Chart I-13Falling Labor Income In Japan Falling Labor Income In Japan Falling Labor Income In Japan Chart I-14Japanese CPI Swaps Are Outliers Japanese CPI Swaps Are Outliers Japanese CPI Swaps Are Outliers A More Complex Picture For The Euro As investors have become more comfortable with the economic and political prospects of the euro area, the euro has become increasingly over-owned, but most importantly, has completely deviated from interest rate parity relationship (Chart I-15). At first glance, this would indicate the euro is greatly vulnerable. This reality, along with very long positioning of speculators in EUR/USD, highlights that it will be difficult for the euro to stay above 1.20 in the coming months. However, for the euro to move below 1.15, U.S. inflation has to pick up. Thus, for the remainder of the year, the EUR/USD is likely to remain range bound between these two numbers. Two factors make the picture less clear for EUR/USD than for USD/JPY. First, the European Central Bank is intent on beginning to taper its asset purchases this year, a move that will be announced in October. At yesterday's press conference, ECB President Mario Draghi was unequivocal about this, despite the slight curtailments to the central bank's inflation forecasts. Moreover, the seeming lack of concern vis-à-vis this year's 6% increase in the trade-weighted euro was perceived by investors as a green light to keep betting on a stronger EUR/USD. Second, as we argued five months ago, exchange rate dynamics are more a function of assets' expected returns than just interest rate differentials.2 As Chart I-16 illustrates, when a portfolio of eurozone stocks, bonds and cash outperforms a similar U.S. one, this leads to a durable rally in EUR/USD. Today, the relative performance of this European portfolio is toward the bottom of its historical distribution, and may even be already turning the corner. If this move has durability, inflows into the euro area could push EUR/USD back into the 1.3 to 1.4 range. Chart I-15Euro Is Expensive ##br##To IRP Euro Is Expensive To IRP Euro Is Expensive To IRP Chart I-16Outperforming Euro Area Assets##br### Could Support EUR/USD Outperforming Euro Area Assets Could Support EUR/USD Outperforming Euro Area Assets Could Support EUR/USD The Loonie Will Keep Flying The Bank of Canada delivered another rate hike this week. The BoC continues to focus on closing the Canadian output gap and the strong economy, ignoring weak wages and inflation. The BoC was rather sanguine regarding the slowdown in real estate activity in Toronto, Canada's largest city, and seemed comfortable with the CAD's recent strength, arguing it was a reflection of Canada's strength and not yet an impediment to it. The CAD interpreted this announcement bullishly. We agree. In a Special Report written last July, we argued that the BoC was among the best-placed central banks to tighten policy among the G10.3 Additionally, the CAD is cheap, trading at a 7% discount to PPP. It is also still below its fair value, implied by interest rate differentials. As such, we continue to overweight the Canadian dollar, being long the loonie against the euro and the Aussie. It also has upside against the USD, but could prove vulnerable to a pick-up in U.S. inflation. Thus, we remain committed to buying the CAD on its crosses. Bottom Line: The euro may be expensive relative to interest rate differentials, but the anticipation around the ECB's tapering continues to represent a support under EUR/USD. As a result, this pair is likely to remain range-bound, roughly between 1.2 and 1.15. USD/JPY has more upside as Japanese inflation expectations and wages are sagging, suggesting the BoJ is nowhere near the ECB in terms of moving away from an ultra-accommodative stance. The CAD will continue to experience upside for the remainder of the year; stay long the loonie on its crosses. Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com 1 Please see The Bank Credit Analyst Monthly Report, "Did Amazon Kill The Phillips Curve?" dated August 3, 2017, available at bca.bcaresearch.com 2 Please see Foreign Exchange Strategy Weekly Report, "The Fed And The Dollar: A Gordian Knot", dated April 14, 2017, available at fes.bcaresearch.com 3 Please see Foreign Exchange Strategy And Global Alpha Sector Strategy Special Report, "Who Hikes Next?", dated June 30, 2017, available at fes.bcaresearch.com Currencies U.S. Dollar Chart II-1USD Technicals 1 USD Technicals 1 USD Technicals 1 Chart II-2USD Technicals 2 USD Technicals 2 USD Technicals 2 The dollar had a particularly eventful week. With Fed officials Brainard and Kashkari unleashing their dovish remarks, the greenback suffered as investors pushed down 10-year yields. While Brainard highlighted her concern over the "recent low readings of inflation", Kashkari took it further and said that the hikes may be "doing real harm" to the economy. Adding to the Fed's concerns, Stanley Fischer, a long-serving Fed official and an ardent supporter of policy normalization, announced his resignation on Wednesday. Mario Draghi's hawkish press conference added further downward pressure on the dollar, with the DXY making a new low of 91.41. It is unlikely that the dollar will be able to meaningfully rally until inflation re-emerges, a year-end event. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Fade North Korea, And Sell The Yen - August 11, 2017 Balance Of Payments Across The G10 - August 4, 2017 The Euro Chart II-3EUR Technicals 1 EUR Technicals 1 EUR Technicals 1 Chart II-4EUR Technicals 2 EUR Technicals 2 EUR Technicals 2 The euro reacted very positively to the ECB monetary policy speech. Draghi highlighted the uncertainty associated with the strong currency, but noted that the ECB doesn't expect it to have a large impact on inflation, which helped the euro hit a high of EUR/USD 1.2018. He nonetheless highlighted that achieving the ECB's price mandate will require "patience" and "persistence" and he expects inflation to hit its target by 2020. Additionally, the ECB lowered its inflation forecasts, while increasing its 2017 growth forecasts. In terms of QE, Draghi clarified that details will be revealed in the next meeting held on October 26, but that interest rates will remain accommodative for an extended period of time. Although President Draghi laid out some concerns about the strong euro, it seems momentum is unlikely to falter unless markets become more positive on the dollar or the pound. We expect this to occur by the end of this year, when inflation picks up again in the U.S. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 The Yen Chart II-5JPY Technicals 1 JPY Technicals 1 JPY Technicals 1 Chart II-6JPY Technicals 2 JPY Technicals 2 JPY Technicals 2 Recent data in Japan has been negative: Industrial production yearly growth declines substantially from June's 5.5% number, coming in at 4.7%. This data point also underperformed expectations. Housing starts contracted by 2.3% on a YoY basis, also underperforming expectations. Meanwhile, labor cash earnings also contracted by 0.3% on a yearly basis, underperforming expectations. As we highlighted a few weeks ago, multiple indicators are signaling a slowdown in the Japanese economy. The recent batch of negative data seems to confirm this view, which means that the dovish bias of the BoJ will only be further reinforced. Consequently the yen will be the mirror image of U.S. bonds. Given that rate expectations have collapsed to the point that the market is only anticipating 30 basis points in hikes in the U.S. over the next 2 years, risks point upwards for USD/JPY. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Fade North Korea, And Sell The Yen - August 11, 2017 Balance Of Payments Across The G10 - August 4, 2017 British Pound Chart II-7GBP Technicals 1 GBP Technicals 1 GBP Technicals 1 Chart II-8GBP Technicals 2 GBP Technicals 2 GBP Technicals 2 Recent data in U.K. has been mixed: Markit manufacturing PMI increased from August to July, coming in at 56.9. This data point also outperformed expectations. Meanwhile, both construction and Markit services PMI underperformed expectations coming in at 51.2 and 53.2, respectively. Finally, nationwide house price year-on-year growth also underperformed expectations, coming in at 2.1%. At the beginning of August, we warned of a repricing of rate expectations in the U.K. given that the pass through from the currency was set to dissipate, while the housing market and real disposable income were undergoing a major slowdown. So far, this view has proven correct, with the pound falling against the dollar and the euro. We expect that GBP/USD has further downside on a 12 month basis, as rate expectations in the U.S. have likely found a bottom. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 Australian Dollar Chart II-9AUD Technicals 1 AUD Technicals 1 AUD Technicals 1 Chart II-10AUD Technicals 2 AUD Technicals 2 AUD Technicals 2 Data out of Australia was not particularly strong: TD Securities Inflation dropped on an annual basis to 2.6% from 2.7%; Gross operating profits contracted at a 4.5% rate, below the expected 4% contraction; Current account balance came in at AUD -9.862 bn, below expectations, following a 59% decrease in the trade balance from the last quarter, and a 4% decrease in the net primary income; Most notably, GDP grew at the expected 1.8% annual rate, albeit faster than the previous growth rate of 1.7%. The RBA decided to leave rates unchanged, but with a slightly hawkish tone. While growth is generally in line with the Bank's forecasts, it was also highlighted that the appreciating exchange rate and low wages remain headwinds for inflation. A brighter housing market was noted as house price increases are slowing down, owing to macroprudential measures. While the Bank sees an improving labor market, we remain skeptical as the underemployment rate has not improved, which is limiting wage growth. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 New Zealand Dollar Chart II-11NZD Technicals 1 NZD Technicals 1 NZD Technicals 1 Chart II-12NZD Technicals 2 NZD Technicals 2 NZD Technicals 2 Surprisingly, in spite of the weakness of the U.S. dollar, the kiwi has been falling for the past month. This has been in part due to some weak data coming out of New Zealand: Building permits continued their decline, with a Month-on-Month decline in July of 0.7%. Both the ANZ Activity Outlook and the Business Confidence indicators declines in August relative to July. The New Zealand terms of trade Index underperformed expectations, coming in at 1.5%. Additionally July's number was revised down from 5.1% to 3.9%. The recent weakness in the NZD might be indicative of some weakness permeating from EM, given that the New Zealand economy is highly sensitive to the global economy. If an EM selloff materializes we expect significant downside for the NZD particularly against the yen. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Who Hikes Next? - June 30, 2017 Canadian Dollar Chart II-13CAD Technicals 1 CAD Technicals 1 CAD Technicals 1 Chart II-14CAD Technicals 2 CAD Technicals 2 CAD Technicals 2 Data has been quite strong out of the Canadian economy recently: The current account deficit was better than expected at CAD -16.32 bn, with the merchandise trade balance also improving; Manufacturing PMI came in at 51.7, beating the expected 51.3; GDP growth came in at an astonishing 4.5% annualized rate; Accordingly, the BoC raised the overnight rate to 1%. Markets were expecting hawkish remarks, but not a hike. The CAD rallied more than 1% against USD on the news, and outperformed all other G10 currencies. Current expectations for a December hike are at 68%, and we agree. The CAD will see further strength against all G10 currencies except USD by the end of the year. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Swiss Franc Chart II-15CHF Technicals 1 CHF Technicals 1 CHF Technicals 1 Chart II-16CHF Technicals 2 CHF Technicals 2 CHF Technicals 2 Recent data in Switzerland has been mixed: Gross Domestic Product yearly growth came in at 0.3%, underperforming expectations and deaccelerating from a month ago. Headline inflation came in line with expectations at 0.5%, it also increase from the previous month reading of 0.3%. Real retails sales underperformed expectations, contracting by 0.7% on a YoY basis. However the SVME PMI outperformed and increased from the July reading, coming in at 61.2. After reaching 1.15 in early August, EUR/CHF has stabilized around to 1.135. Overall the Swiss economy is still too weak for the SNB to change their stance on currency intervention. Core Inflation will have to pick up to at least 1% for the SNB to consider a change in stance and let go of the implied floor in EUR/CHF. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 Norwegian Krone Chart II-17NOK Technicals 1 NOK Technicals 1 NOK Technicals 1 Chart II-18NOK Technicals 2 NOK Technicals 2 NOK Technicals 2 Recent data in Norway has been positive: Retails sales monthly growth came in at 0.4%, recovering from last month negative reading ad outperforming expectations. Manufacturing output growth also outperformed expectations, coming in at 1.5%. Finally registered unemployment came in at 2.7%, declining from last month reading and coming in line with expectations. USD/NOK has continued to go down as rate expectations for the U.S. have decreased and oil prices have increased thanks to the refining shut-downs in Texas due to hurricane Harvey. We expect this trend to reverse once rate expectations in the U.S. start to go up. However, we do expect more downside in EUR/NOK as this cross is much more sensitive to oil prices. Report Links: 10 Charts For A Late-August Day - August 25, 2017 Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Swedish Krona Chart II-19SEK Technicals 1 SEK Technicals 1 SEK Technicals 1 Chart II-20SEK Technicals 2 SEK Technicals 2 SEK Technicals 2 Swedish data was largely downbeat: Retail sales are growing at a 3.7% annual rate, in line with expectations; The Swedish trade balance went into a deficit in July of SEK -0.5 bn from a SEK 5.4 bn surplus in June; Consumer confidence decreased to 100.3 from 102.2 and below the expected 103; Manufacturing PMI also disappointed at 54.7, below the expected 60; Swedish IP is growing at a still high 5.3% annual pace, but less than the previous 8.9% rate; While this data was somewhat weak, Swedish inflation is at or above its target across all measures. The Riksbank left its repo rate unchanged at -0.5%. In its press release, the Bank highlighted high growth and inflation but stated that the rate will not be increased until the middle of 2018. It also increased inflation forecasts, with CPI and CPIF predicted at 2.9% and 2.1% by 2019. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Bloody Potomac - May 19, 2017 Trades & Forecasts Forecast Summary Core Portfolio Closed Trades

Feature Healthy consumer spending driving a booming sales environment, along with the operating leverage that high revenue growth produces, have been the key underpinnings of the nascent revival in the S&P 500 margin expansion. This has occurred against the backdrop of muted wage growth in most sectors which has amplified margin expansion. We recently showed that S&P 500 operating leverage has historically added $1.4 of earnings for every $1 of incremental revenues (please see our Weekly Report of April 17, 2017 for more details). On a trailing 12-month basis, the S&P 500 has added more than $3 of earnings for every $1 of incremental revenues, more than double the historical average operating leverage. Clearly this pace of margin expansion is unsustainable, particularly since the tight labor market seems likely to force a reacceleration in wage growth. A common narrative among investors has been that late-cycle dynamics will soon force a mean reversion in S&P 500 operating margins. However, and while every economic cycle is different, true mean reversion only happens in recessions (Chart 1). Chart 1Margins Can Expand From Here Margins Can Expand From Here Margins Can Expand From Here Further, the absolute margin level of the S&P 500 is far from being without precedent. Since the 1970's, margins have typically peaked for the cycle only after approaching one standard deviation above the trend and the current S&P500 margin is just past halfway there. It is also worth noting that margins can stay extended for a considerable time; margins have surpassed one standard deviation above trend twice this decade without a material retrenchment. Chart 2 shows the high, low and current trailing operating margins of the S&P 500 and the eleven GICS1 sectors. At first glance, it appears that margins are particularly high in the heavyweight financials and IT sectors. Some context is required; both sectors experienced bubbles in the last two decades that saw operating profits plumb extreme lows in the subsequent busts, making their profit ranges appear unusually broad. Chart 3 corrects to exclude two-standard deviation events for all sectors. The message is clear: margins still have significant room to run. Chart 2High, Low And Current Trailing S&P 500 Operating Margins Sector Margin Outlook: Profit Growth Is Not Done Yet Sector Margin Outlook: Profit Growth Is Not Done Yet Chart 3High, Low And Current Trailing S&P 500 Operating Margins, Normalized Sector Margin Outlook: Profit Growth Is Not Done Yet Sector Margin Outlook: Profit Growth Is Not Done Yet Operating margins in isolation only tell part of the story. In Chart 4, we compare profitability to the capital deployed in pursuit of said profits. Capital deployed and its earned return should theoretically plot on a linear function; plotting above the fitted regression line implies insufficient returns, while plotting below the line indicates excess returns. In our analysis, most sectors plot relatively closely to the market line with a few notable outliers. Financials are likely earning significant excess returns on capital, while utilities are waving a warning flag. We reiterate our overweight and underweight ratings on these two sectors, respectively (Chart 4). Chart 4Capital Intensity Of Profits Sector Margin Outlook: Profit Growth Is Not Done Yet Sector Margin Outlook: Profit Growth Is Not Done Yet The upshot of high margins and low capital requirements is above-average return on capital. Consequently, rising valuation multiples move in tandem with ROIC and vice-versa. Our analysis bears that out; financials are relatively far along the continuum along which most of the S&P 500 sectors plot, though still modestly below the fitted regression line indicating fair value. Conversely, real estate, while attractive from a return on capital perspective, is highly overvalued (Chart 5). Chart 5Margin Efficiency And Valuation Sector Margin Outlook: Profit Growth Is Not Done Yet Sector Margin Outlook: Profit Growth Is Not Done Yet This Special Report takes a sector-by-sector view on the margin outlook that supports our thesis of ongoing margin gains delivering an earnings-driven stock market rally. Chris Bowes, Associate Editor U.S. Equity Strategy chrisb@bcaresearch.com Chart 6Oil Stocks Look Set To Decline Oil Stocks Look Set To Decline Oil Stocks Look Set To Decline Chart 7Capital Formation Should Take Off Capital Formation Should Take Off Capital Formation Should Take Off Chart 8Consumers Have Opened Their Wallets Consumers Have Opened Their Wallets Consumers Have Opened Their Wallets Chart 9Surging Global Manufacturing Surging Global Manufacturing Surging Global Manufacturing Chart 10Real Estate Rents Look##br## Set To Decline Real Estate Rents Look Set To Decline Real Estate Rents Look Set To Decline Chart 11The Right Conditions For Industrial##br## Margin Expansion The Right Conditions For Industrial Margin Expansion The Right Conditions For Industrial Margin Expansion Chart 12Dark Clouds On The Horizon ##br##For Health Care Margins Dark Clouds On The Horizon For Health Care Margins Dark Clouds On The Horizon For Health Care Margins Chart 13Utilities Margins Are##br## Likely To Contract Utilities Margins Are Likely To Contract Utilities Margins Are Likely To Contract S&P Energy (Overweight) Chart 14S&P Energy S&P Energy S&P Energy Energy operating profit margins have been on a wild ride, collapsing with the underlying commodity and then partially recovering as the industry rationalized. Analysts are forecasting more of the same, with the industry forecast to generate profits for the first time in more than two years. Pricing power has spiked higher, though from an extremely low base, as the aforementioned industry rationalization has taken hold. Wage growth looks fairly tepid and the net margin impact supports the forecast view of margin expansion. Rampant cost inflation appears to be a thing of the past. Accordingly, the essential component for margin recovery will be top line growth. The key factors in a top-line growth scenario for the energy sector will be a demand-driven recovery in crude oil prices, supported by continued supply-side discipline. The current global economic revival and pause in the U.S. dollar bull market are catalysts for the former while OPEC 2.0 supply cuts (with effective compliance) and lower crude supply are catalysts for the latter. Encouragingly, the rig count remains well below peak levels, Cushing crude oil inventories are contracting on a year-over-year basis and OECD oil stocks appear poised to contract in late autumn/early winter (Chart 6). Net, we are constructive on energy sector margins (Chart 14). S&P Financials (Overweight) Chart 15S&P Financials S&P Financials S&P Financials Margins, though below historic peak levels, have improved dramatically. The stock market has not rewarded the sector for the solid performance, making financials a standout sector where earnings have led prices higher, rather than multiple expansion. A healthy consumer, housing market and corporate sector should lead to strong capital formation which, in turn, implies improving revenue growth for financials. This is captured by our loans & leases model which points to the largest upswing in credit growth of the past 30 years (Chart 7) Banks in particular benefit from a healthy economy as very low unemployment should be accompanied by solid loan quality which makes the industry's margin gains more durable (Chart 7). We expect banks, as the largest segment of the financials sector, to lead the index higher. Pricing power and wage growth have recently been diverging with the former moving steeply positive and the latter falling to the slowest growth of the past 5 years. These moves bode well for future margin expansion; analysts agree, with forecasts pointing to margins approaching twenty-year highs (Chart 15). S&P Consumer Discretionary (Overweight) Chart 16S&P Consumer Discretionary S&P Consumer Discretionary S&P Consumer Discretionary Consumer discretionary margins have inflated dramatically and, despite a moderation in actual and forecast profitability, they remain more than one standard deviation above normal. Wage growth is declining from fairly eye-watering levels but still remains faster than the muted sector pricing power. The net of these points is falling margins, in line with analyst forecasts. Spending has recently poked higher as a much improved household balance sheet and wage growth have made the consumer feel flush enough to start spending some of their accumulated savings of the past few years (Chart 8). This resurgence in demand should mean, barring any external shock, that pricing power will recover, though a tight labor market could present a considerable offset via above-normal wage growth. Within the index, margin strength is particularly notable in Home Improvement Retail and Cable & Satellite; both are benefitting from the themes noted above and have seen revenue growth driving wider margins. The Auto Components index is a rare underperformer with margins shrinking as the companies adjust to slowing North American light vehicle production. Net, we remain positive on consumer discretionary profit growth (Chart 16). S&P Consumer Staples (Overweight) Chart 17S&P Consumer Staples S&P Consumer Staples S&P Consumer Staples Consumer staples margins have seen a general upward trajectory over the past three years, though have recently rolled over. The key culprits have been food & drug deflation with retail struggling to maintain profits. Forecasts are pointing to a resumption of the upward margin trend, in line with our improving proxy measure (Chart 17, bottom panel). Eventually staples will regain some share of the consumer's wallet. The wage bill is moving in the right direction and even a modest uptick in sector pricing power could trigger margin expansion. It is worth noting that consumer staples is our only remaining overweight defensive index as we have drifted toward cyclical sectors with our increasingly bullish stance over the course of the year. Still, we remain confident of a modest sector margin recovery, though expect consumer discretionary to have a better profit growth profile. S&P Telecommunication Services (Neutral) Chart 18S&P Telecom Services S&P Telecom Services S&P Telecom Services S&P telecom services is at the very bottom of the GICS1 sector EPS growth table this year despite easy comparable quarters in 2016; this is reflected in the index's steady downward drift (Chart 18, top panel). Still, margins have started staging a recovery and the sell-side appears reasonably optimistic. The issue is pricing, the weakness of which is taking profits down regardless of margin resilience. Encouragingly, selling prices cannot contract at 10% per annum indefinitely and recent anecdotal evidence from earnings calls suggests that the peak deflationary impulse is likely behind the industry. Impressive labor cost discipline along with even a modest pricing power rebound signal that a grinding higher margin backdrop is likely in the coming months, though our margin proxy is weighed down by still-falling pricing power (Chart 18, bottom panel). S&P Materials (Neutral) Chart 19S&P Materials S&P Materials S&P Materials Margins in the S&P materials index have recovered sharply from their recent lows, with analysts forecasting continued margin expansion. Said margin expansion will be dependent on the industry holding on to the pricing power gains it has made over the past year; we think odds are good this can happen. A global manufacturing rebound appears to be underway; the global manufacturing PMI has recently reaccelerated and jumped to a six year high (Chart 9). Further, it looks likely that a coordinated central bank tightening cycle has begun which should make U.S. exports relatively more attractive, even if the greenback moves laterally from current levels. With respect to chemicals, the dominant materials component industry, a wave of global mergers (Chart 9) should limit price competition while also stripping out some overcapacity which has been a perennial margin overhang. As well, domestic operating conditions have taken a turn for the better as U.S. chemical production has troughed and utilization rates have improved (Chart 9). Still, inventories have surged in advance of the manufacturing recovery (not shown) and any demand misstep could have serious margin implications. Our materials margin proxy points to modest margin gains (Chart 19). S&P Real Estate (Neutral) Chart 20S&P Real Estate S&P Real Estate S&P Real Estate The S&P Real Estate index comprises mostly REITs and does not compare well to the other sectors on an operating margin basis, owing to the vastly different business model. Still, a discussion of drivers of both revenues and costs is worthwhile. Real estate occupancy rates have crested and generationally high supply additions in the apartment space are all but certain to push vacancies higher still (Chart 10). The implication is that rental inflation will remain under intense downward pressure, as has been the case since the beginning of 2016. Worrisomely, credit quality in select commercial real estate (CRE) segments is deteriorating at the margin (Chart 10). Should the trend worsen, REIT margins will deteriorate. According to a recent Fed Senior Loan Officer Survey, bankers are less willing to extend CRE credit. In fact, if one excludes the GFC spike, the tightening in CRE lending standards is near the two previous recessionary highs (Chart 10 on page 8). If banks continue to close the credit taps, CRE prices will suffer a setback. Nevertheless, the tight labor market and accelerating industrial production should keep the appetite for CRE upbeat and prices may have a bit more room to run before reaching a cyclical peak (Chart 20). S&P Industrials (Neutral) Chart 21S&P Industrials S&P Industrials S&P Industrials A demand revival, both domestic and globally, has helped drive a recovery of S&P industrials margins from the mini manufacturing recession of 2015/early-2016. The U.S. dollar bull market has paused (Chart 11), global demand and credit growth has recovered (Chart 11) and domestic optimism abounds (Chart 11); all the conditions look supportive of the consistent margin profile forecast by the sell-side. However, the margin expansion thesis is not without risk; pricing power gains appear to have rolled over while the wage bill, the weakness of which was a significant margin driver, has spiked. The result is that our industrials margin proxy has eased, though we discount the measure as it has not correlated well with observed margins. Still, if demand continues to remain upbeat, the operating leverage impact on the relatively high fixed cost sector should offset labor cost spikes. Net, we expect margins to drift mostly sideways (Chart 21). S&P Health Care (Underweight) Chart 22S&P Health Care S&P Health Care S&P Health Care S&P health care margins are showing warning signs of a potential retreat. Pricing power has worsened significantly since recent highs in 2016 which could warn of a top line contraction, particularly in the context of drug price inflation. Chart 12 shows that since 2005 drug prices have nearly doubled and the slope has actually steepened since 2011. Health care spending in the U.S. comprises over 17% of GDP, the highest in the world, but it has likely plateaued. Real health care spending is decelerating in absolute terms, and had been contracting compared with overall PCE earlier this year (Chart 12). This suggests that selling price blues are demand driven and will likely continue to weigh on health care profits. Not only are selling prices softening, but also the health care sector wage bill is on fire, pushing multi-year highs. Taken together, operating margins will continue to compress, sustaining the recent down drift. Should margins worsen as we expect, the recent updraft in the index price should follow earnings downward (Chart 22). S&P Utilities (Underweight) Chart 23S&P Utilities S&P Utilities S&P Utilities In earlier sections of this report, we have discussed the beneficiaries of growing ebullience in global economic expectations; utilities are at the opposite end of the spectrum. Now that the Fed is ready to start unwinding its balance sheet, the ECB is preparing the waters for QE tapering and a slew of CBs are on the cusp of a new tightening interest rate cycle, there are high odds that fixed income proxies, utilities among them, will continue to suffer. From a profit perspective, our margin proxy is pointing to a pricing driven recovery. However, contracting natural gas prices, the marginal price setter for the industry, suggest that recent utilities pricing power gains are running on empty (Chart 13). Tack on waning productivity, with labor additions handily outpacing electricity production, and the ingredients for a margin squeeze are in place (Chart 13). Importantly, industry utilization rates are probing multi-decade lows and overcapacity is negative for pricing power. Chart 13 confirms that utilities construction is relentless at a time when turbine and generator inventories have been hitting all-time highs. This is a deflationary backdrop, and suggests that sell-side analyst optimism is wrong footed. Net, we think margin weakness should persist (Chart 23). S&P Information Technology (Underweight) Chart 24S&P Information Technology S&P Information Technology S&P Information Technology Margins in the S&P information technology index are pushing their 20-year highs. However, the sector is a story of leaders and laggards. The technology hardware, storage & peripherals sub-index (almost entirely AAPL), for example, has seen their operating margin roughly double in the past ten years. Conversely, communications equipment is in the midst of a collapse in pricing power as intense competition has engulfed telcos (their principal customer group) and the uncertainty in the federal government has held back outlays. Our margin proxy is pointing to a modest margin contraction, a result of slipping sector pricing power partially offset by a flat to slightly negative sector wage bill. This stands in contrast to sell-side forecasts who expect margins to hit record levels in the next year. We view the sell-side as overly sanguine with respect to margins and expect pricing power to weigh in coming months (Chart 24).

Highlights Some caution warranted here. Hurricane Harvey's impact on the economy and markets. Tensions in North Korea will linger. NIPA and S&P now telling same story on profits, margins. Is the August employment report enough for the Fed? Feature The impact of Hurricane Harvey will ripple through the economic data in the coming months, but will not impact the overall trajectory of the economy or the Fed. However, elevated equity valuations, escalating tensions in North Korea, a widening disconnect between the bond market and the Fed and profit growth that is poised to peak in the second half of the year warrants careful attention from investors. Nonetheless, we remain slightly overweight stocks and favor stocks over bonds. Caution On Risk Assets We recommend that clients be prudent, paring back any overweight positions and holding some safe-haven assets within diversified portfolios. BCA research has demonstrated that U.S. Treasuries, Swiss bonds and JGBs were the best performers during a crisis (Chart 1). The same is true for the Swiss franc and the Japanese yen, such that the currency exposure should not be hedged in these cases. The dollar is more nuanced. It tends to perform well during financial crises, but not in geopolitical crises or recessions. Chart 1Gold Loves Geopolitical Crises Shelter From The Storm Shelter From The Storm Gold tends to perform well in geopolitical events, although not in recessions or financial crunches. Our base case projects stocks outperforming cash and bonds over the next 6-12 months. BCA's dollar and duration positions have disappointed so far this year. Much hinges on U.S. inflation. Investors appear to have adopted the stance that structural headwinds to inflation will forever dominate the cyclical pressures. Therefore, the bond market is totally unprepared for any upside shocks on the inflation landscape. Admittedly, a rise in bond yields may not be imminent, but the risks appear to be predominantly to the upside. Harvey's Lingering Aftermath History shows that natural disasters such as Hurricane Harvey have a temporary effect on the U.S. economy, the financial markets and the Fed. Ultimately, the macro environment in place before the storm will reassert itself. Nonetheless, it may be a few months before investors determine the long-term impact of the record rainfall and flooding in Houston. Chart 2 shows the ranking of Harvey's preliminary damage estimate of $30B versus other storms of similar magnitude. We are still several weeks away from the peak of the Atlantic hurricane season (mid-September) and two of the most destructive storms in the past 25 years made landfall in mid-to-late October (Wilma and Sandy). Chart 2Economic Impact From Major Hurricanes Economic Impact From Major Hurricanes Economic Impact From Major Hurricanes Chart 3 shows the performance of key economic, inflation and financial market indicators in the past two years and also around five major hurricanes since 1992. Most of the activity-related economic statistics are volatile in the aftermath of the storms and then they recover. The Citi economic surprise index initially moves higher after a storm, and then fades (Chart 3A). There are big swings in housing starts and industrial production and employment growth slows. Inflation tends to climb post-landfall (Chart 3B). In prior episodes, core PCE and core CPI have accelerated along with gasoline prices. Consumer confidence dips initially, but then recovers. Wages are volatile, but tend to accelerate after several months. Chart 3C shows that stocks drift lower for several months following hurricanes and subsequently recoup the losses. The stock-to-bond ratio also moved lower, but regains its pre-storm heights about two months later. Treasury yields fall after storms, but we note that yields have been in a secular decline for 25 years. Chart 3AMajor Hurricane Impact##BR##On Activity Data Major Hurricane Impact On Activity Data Major Hurricane Impact On Activity Data Chart 3BMajor Hurricane Impact On##BR##Sentiment And Inflation Data Major Hurricane Impact On Sentiment And Inflation Data Major Hurricane Impact On Sentiment And Inflation Data Chart 3CMajor Hurricane Impact On##BR##Financial Markets & The Fed Major Hurricane Impact On Financial Markets & The Fed Major Hurricane Impact On Financial Markets & The Fed Hurricane Harvey will not shake the Fed. Nonetheless, the central bank will acknowledge the disaster in the FOMC statement, the FOMC minutes, and/or in Fed Chair Janet Yellen's news conference. We are unchanged in our view that policymakers will begin to pare its balance sheet later this month and bump up rates again in December, assuming that core inflation shows some signs of strength between now and then. History shows (Chart 3C) that, on average, the Fed funds rate tends to move higher in the months after storms hit, but the primary message is that the Fed just continues to do whatever it was doing before the storm. The Fed cut rates in the aftermath of Hurricane Andrew in 1992 in what turned out to be the final rate reduction of the cycle that began in 1989. Ivan hit in September 2004, but the monetary authority raised rates in the final three FOMC meetings of 2004, including at the meeting only a week after the hurricane made landfall. Similarly, the Fed clung to its rate hike regime after Wilma in October 2005. In 2008, Ike arrived in Texas two days before Lehman Brothers collapsed in mid-September. The Fed, which had been cutting rates since September 2007, lowered rates in the final months of 2008. The Fed announced QE3 in late summer 2012 and continued with the program after Sandy came ashore at the end of October 2012. Harvey will be a game changer in some respects: the devastation reduces the odds of a government shutdown or of failing to increase the debt ceiling. We have maintained that there were extremely low odds that the debt ceiling would not be raised. We stated that there was a 25% chance of a government shutdown between October 1, when the current funding expires, and sometime in mid-October when the debt ceiling will hit according to the Congressional Budget Office. However, it would be unfathomable to shut down the government and force the Federal Emergency Management Agency (FEMA) to cease operations. The resulting outrage would damage the Republicans, especially in Texas. Bottom Line: Harvey may have a near-term impact on the economy, but the Fed will stick to its plan. The catastrophe makes it increasingly likely that the debt ceiling will be raised and a resolution will be passed to keep the government operating into the new fiscal year. Thus, equity investors can safely ignore these two risks, and focus on the key risk in the outlook: North Korea. North Korea Could Linger Over Markets BCA believes that the probability of a war on the Korean Peninsula is very low,1 but it may take a while before the uncertainty in Northeast Asia is resolved. Between now (escalating tensions) and then (a negotiated settlement), there will be more provocations and market volatility. There are long-standing constraints to war. The first is a potentially high death toll: Pyongyang can inflict massive civilian casualties in Seoul with a conventional artillery barrage. Furthermore, U.S. troops, and Japanese forces and civilians, would also suffer. Secondly, China is unlikely to remain neutral. Strategically, China will not tolerate a U.S. presence on its border with North Korea. Nevertheless, Washington must establish a credible threat of military action if it is to convince Pyongyang that negotiations offer a superior outcome. It is unclear how long it will take Trump to convince North Korea that the threat of a U.S. preemptive strike is credible. Chart 4 shows the arc of diplomacy2 that the U.S. took with Iran between 2010 and 2014. From an investor perspective, it will be difficult to gauge whether the brinkmanship and military posturing are part of this territorial threat display or evidence of real preparations for an actual attack. More market volatility may occur, but for the time being, we do not think that the tensions in the Korean peninsula will end the bull market in global equities. Positions in traditional safe-haven assets, such as gold, U.S. Treasuries, Swiss francs and (perhaps) Japanese yen, should be considered as hedges against increased market swings. Chart 4Arc Of Diplomacy: Tensions Ramp Up As Nuclear Negotiations Begin Shelter From The Storm Shelter From The Storm Update: Equity Valuations, Sentiment And Technicals U.S. equity valuations are stretched, but elevated valuations alone are not enough to prompt a sell-off in stocks. The BCA valuation indicator is in overvalued territory, where it has been since late 2013. History shows3 that stocks can stay overvalued for extended periods, even when the Fed is raising rates, but policy is still accommodative as it is today. BCA's composite valuation indicator is still shy of the +1 standard deviation level that defines extremely over-valued (Chart 5). However, this is due to the components that compare equity prices with bond yields. The other three elements of the equity indicator, which are unrelated to bond yields, suggest that stock valuation is stretched (Chart 5 panels 2, 3 and 4). That said, equities are attractively priced relative to competing assets, such as corporate bonds and Treasuries (Chart 6). Chart 5U.S. Equities##BR##Are Overvalued... U.S. Equities Are Overvalued... U.S. Equities Are Overvalued... Chart 6...But Look Less Expensive##BR##Relative To Competing Assets ...But Look Less Expensive Relative To Competing Assets ...But Look Less Expensive Relative To Competing Assets Valuation is not a reliable tool to time market turning points and, absent a significant deterioration in the economic, profit and margin environment, we do not forecast a sustained pullback in stocks. Looking beyond BCA's tactical 6-12 month window, above-average market multiples alone imply below-average returns for stocks across a strategic time horizon. Chart 7No Strong Signal From##BR##Sentiment Or Technicals No Strong Signal From Sentiment Or Technicals No Strong Signal From Sentiment Or Technicals BCA's technical and sentiment indicators are not at extremes (Chart 7). The BCA technical indicator, while above zero, is not at a level that in the past has triggered a stock pullback. Similarly, the BCA investor sentiment composite index, while at the top end of its bull market range, is not at an extreme. Moreover, only 50% of the stocks in the NYSE composite are above their 10-week moving average, a level which has not been previously associated with major equity sell-offs. Bottom Line: The solid earnings backdrop remains in place for U.S. stocks as measured by either the S&P or the national accounts. We anticipate that profit growth has peaked according to S&P 500 data on a 4-quarter moving total basis due to tough comparisons although it will slip only modestly in the second half of the year. Next year will see EPS growth drop back into the mid-single digit range. The consensus estimate for 2018 EPS growth is 11%. While valuations are elevated, neither sentiment nor technical indicators are flashing red. We recommend stocks over bonds in the next 6-12 months, but acknowledge that risks to BCA's stance are climbing. A Reconnection In Q2 The Q2 data show that the NIPA and S&P earnings measures have reconnected. In our July 3, 2017 Weekly Report "Summer Stress Out"4 we highlighted the apparent disconnect between the S&P and NIPA, sales earnings and margin data through Q1 2017. The release of the Q2 corporate profits data in the national accounts and the end the Q2 S&P 500 reporting season allow us to provide an update. The year-over-year reading on the NIPA earnings measure ticked up in Q2 while the S&P-based metric ticked down. That said, while there are marked differences in annual growth rates between the two measures, the levels were close to the same point in the second quarter of 2017 (Chart 8, bottom panel). Chart 9 shows that a wide difference persists between corporate sales measured by S&P and the national accounts. Margins calculated on the S&P basis climbed in Q2 while NIPA margins held steady. Even so, a modest gap still remains between NIPA margins at 15.2% and S&P margins at 13.2%. Most of the divergence is related to the denominator of the calculation. The NIPA denominator is corporate sector Gross Domestic Product (GDP). This is a value-added concept that is different from sales. It is not clear why, but GDP has grown much faster than sales since the end of 2014. Chart 8S&P And NIPA##BR##Profit Comparison S&P And NIPA Profit Comparison S&P And NIPA Profit Comparison Chart 9Denominator Explains##BR##S&P/NIPA Margin Divergence Denominator Explains S&P/NIPA Margin Divergence Denominator Explains S&P/NIPA Margin Divergence We believe that the S&P statistics are painting a more accurate picture because sales are easier to measure while value-added is more complicated. The slow growth of sales is not a bullish point for stocks. Nonetheless, it does not appear that financial engineering has distorted bottom-up company data to such an extent that the S&P readings are falsely signaling strong profit growth. We expect the secular mean-reversion of margins to re-assert itself in the S&P data, perhaps beginning early in 2018. Nonetheless, the profit backdrop is positive for stocks for now. Is The August Jobs Report Enough For The Fed? Chart 10Labor Market Conditions##BR##Favor Risk Assets Labor Market Conditions Favor Risk Assets Labor Market Conditions Favor Risk Assets U.S. payrolls expanded by 156,000 in August. Relative to the underlying growth rate in the labor force, this is still a healthy pace of jobs growth. Nevertheless, it fell short of expectations for a 180,000 increase and the prior two months saw a cumulative downward revision of 41,000. The August data were not impacted by Hurricane Harvey. Aggregate hours worked, a measure of total labor inputs based on changes in employment and the workweek, fell by 0.2% m/m. That said, aggregate hours worked are up 1.3% at a quarterly annualized rate thus far in Q3. This is consistent with GDP growth of a bit over 2%, which has been the trend in the current economic expansion. Meanwhile, wage gains remain muted. Average hourly earnings rose just 0.1% m/m. Annual wage inflation has been steady at 2.5% for several months now (Chart 10, bottom panel). If productivity is expanding modestly around 1%, the current pace of wage gains would suggest that unit labor costs are growing around 1.5%. This will make it difficult for general price inflation to accelerate to the Fed 2% target. Nonetheless, the reacceleration in the 3-month change in average hourly earnings from 1.9% in January 2017 to 2.6% in August supports the Fed's view on inflation. Finally, the unemployment rate ticked up to 4.4% from 4.3%. This was because the separate household survey showed a 74,000 drop in employment. The participation rate held steady at 62.9% in August. Bottom Line: While falling short of expectations in August, U.S. employment growth remains solid and job gains are continuing at a pace consistent with the 2% GDP growth rate of recent years. However, muted wage gains mean that progress to the Fed's 2% inflation target is looking suspect. We anticipate that the Fed will announce the process of running down its balance sheet at the September FOMC meeting. Rate hikes are on hold at least until the December FOMC meeting, and even then only if core inflation shows some signs of strength in the next few months. U.S. risk assets should continue to benefit from moderate growth, low inflation and a "go slow" approach by the Fed. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com 1 Please see BCA's Geopolitical Strategy Weekly Report, "Can Pyongyang Derail The Bull Market?", August 16, 2017. It is available at gps.bcaresearch.com. 2 Please see BCA's Geopolitical Strategy Weekly Report, "Northeast Asia: Moonshine, Militarism, And Markets? ,"May 24, 2017, available at gps.bcaresearch.com. 3 Please see BCA's U.S. Investment Strategy Weekly Report, "Sizing Up The Second Half", July 10, 2017, available at usis.bcaresearch.com. 4 Please see BCA's U.S. Investment Strategy Weekly Report, "Summer Stress Out", July 3, 2017, available at usis.bcaresearch.com.

Dear Client, In addition to this Special Report written by my colleague Mark McClellan, we are sending you an abbreviated weekly report, which includes the Tactical Global Asset Allocation Monthly Update. Best regards, Peter Berezin, Chief Global Strategist Global Investment Strategy Highlights A "culture of profound cost reduction" has gripped the business sector since the GFC according to one school of thought, permanently changing the relationship between labor market slack and wages or inflation. If true, it could mean that central banks are almost powerless to reach their inflation targets. Amazon, Airbnb, Uber, robotics, contract workers, artificial intelligence, horizontal drilling and driverless cars are just a few examples of companies and technologies that are cutting costs and depressing prices and wages. In the first of our series on inflation, we will focus on the rise of e-commerce and the related "Amazonification" of the economy. In theory, positive supply shocks should not have more than a temporary impact on inflation if the price level is indeed a monetary phenomenon in the long term. But a series of positive supply shocks could make it appear for quite a while that low inflation is structural in nature. We are keeping an open mind and reserving judgement on the disinflationary impact of robotics, artificial intelligence and the gig economy until we do more research. But in terms of the impact of e-commerce, it is difficult to find supportive evidence at the macro level. The admittedly inadequate measures of online prices available today do not suggest that e-commerce sales are depressing the overall inflation rate by more than 0.1 or 0.2 percentage points. Moreover, it does not appear that the disinflationary impact of competition in the retail sector has intensified over the years. Today's creative destruction in retail may be no more deflationary than the shift to 'big box' stores in the 1990s. Perhaps lower online prices are forcing traditional retailers match the e-commerce vendors, allowing for a larger disinflationary effect than we estimate. However, the fact that retail margins are near secular highs outside of department stores argues against this thesis. The sectors potentially affected by e-commerce make up a small part of the CPI index. The deceleration of inflation since the GFC has been in areas unaffected by online sales. High profit margins for the overall corporate sector and depressed productivity growth also argue against the idea that e-commerce represents a large positive macro supply shock. Perhaps the main way that e-commerce is affecting the macro economy and financial markets is not through inflation, but via the reduction in the economy's capital spending requirement. This would reduce the equilibrium level of interest rates, since the Fed has to stimulate other parts of the economy to offset the loss of demand in capital spending in the retail sector. Feature Anecdotal evidence is all around us. The global economy is evolving and it seems that all of the major changes are deflationary. Amazon, Airbnb, Uber, robotics, contract workers, artificial intelligence, horizontal drilling and driverless cars are just a few examples of companies and technologies that are cutting costs and depressing prices and wages. Central banks in the major advanced economies are having difficulty meeting their inflation targets, even in the U.S. where the labor market is tight by historical standards. Based on the depressed level of bond yields, it appears that the majority of investors believe that inflation headwinds will remain formidable for a long time. One school of thought is that low inflation reflects a lack of demand growth in the post-Great Financial Crisis (GFC) period. Another school points to the supply side of the economy. A recent report by Prudential Financial highlights "...obvious examples of ... new business models and new organizational structures, whereby higher-cost traditional methods of production, transportation, and distribution are displaced by more nontraditional cost-effective ways of conducting business." 1 A "culture of profound cost reduction" has gripped the business sector since the GFC according to this school, permanently changing the relationship between labor market slack and wages or inflation (i.e., the Phillips Curve). Employees are less aggressive in their wage demands in a world where robots are threatening humans in a broadening array of industrial categories. Many feel lucky just to have a job. In a highly sensationalized article called "How The Internet Economy Killed Inflation," Forbes argued that "the internet has reduced many of the traditional barriers to entry that protect companies from competition and created a race to the bottom for prices in a number of categories." Forbes believes that new technologies are placing downward pressure on inflation by depressing wages, increasing productivity and encouraging competition. There are many factors that have the potential to weigh on prices, but analysts are mainly focusing on e-commerce, robotics, artificial intelligence, and the gig economy. In the first of our series on inflation, we will focus on the rise of e-commerce and the related "Amazonification" of the economy. The latter refers to the advent of new business models that cut out layers of middlemen between producers and consumers. Amazonification E-commerce has grown at a compound annual rate of more than 9% over the past 15 years, and now accounts for about 8½% of total U.S. retail sales (Chart 1). Amazon has been leading the charge, accounting for 43% of all online sales in 2016 (Chart 2). Amazon's business model not only cuts costs by eliminating middlemen and (until recently) avoiding expensive showrooms, but it also provides a platform for improved price discovery on an extremely broad array of goods. In 2013, Amazon carried 230 million items for sale in the United States, nearly 30 times the number sold by Walmart, one of the largest retailers in the world. Chart 1E-Commerce: Steady Increase In Market Share E-Commerce: Steady Increase In Market Share E-Commerce: Steady Increase In Market Share Chart 2Amazon Dominates Did Amazon Kill The Phillips Curve? Did Amazon Kill The Phillips Curve? With the use of a smartphone, consumers can check the price of an item on Amazon while shopping in a physical store. Studies show that it does not require a large price gap for shoppers to buy online rather than in-store. Amazon appears to be impacting other retailers' ability to pass though cost increases, leading to a rash of retail outlet closings. Sears alone announced the closure of 300 retail outlets this year. The devastation that Amazon inflicted on the book industry is well known. It is no wonder then, that Amazon's purchase of Whole Foods Market, a grocery chain, sent shivers down the spines of CEOs not only in the food industry, but in the broader retail industry as well. What would prevent Amazon from applying its model to furniture and appliances, electronics or drugstores? It seems that no retail space is safe. A Little Theory Before we turn to the evidence, let's review the macro theory related to positive supply shocks. The internet could be lowering prices by moving product markets toward the "perfect competition" model. The internet trims search costs, improves price transparency and reduces barriers to entry. The internet also allows for shorter supply chains, as layers of wholesalers and other intermediaries are removed and e-commerce companies allow more direct contact between consumers and producers. Fewer inventories and a smaller "brick and mortar" infrastructure take additional costs out of the system. Economic theory suggests that the result of this positive supply shock will be greater product market competition, increased productivity and reduced profitability. In the long run, workers should benefit from the productivity boost via real wage gains (even if nominal wage growth is lackluster). Workers may lower their reservation wage if they feel that increased competitive pressures or technology threaten their jobs. The internet is also likely to improve job matching between the unemployed and available vacancies, which should lead to a fall in the full-employment level of unemployment (NAIRU). Nonetheless, the internet should not have a permanent impact on inflation. The lower level of NAIRU and the direct effects of the internet on consumer prices discussed above allow inflation to fall below the central bank's target. The bank responds by lowering interest rates, stimulating demand and thereby driving unemployment down to the new lower level of NAIRU. Over time, inflation will drift back up toward target. In other words, a greater degree of the competition should boost the supply side of the economy and lower NAIRU, but it should not result in a permanently lower rate of inflation if inflation is indeed a monetary phenomenon and central banks strive to meet their targets. Still, one could imagine a series of supply shocks that are spread out over time, with each having a temporary negative impact on prices such that it appears for a while that inflation has been permanently depressed. This could be an accurate description of the current situation in the U.S. and some of the other major countries. We have sympathy for the view that the internet and new business models are increasing competition, cutting costs and thereby limiting price increases in some areas. But is there any hard evidence? Is the competitive effect that large, and is it any more intense than in the past? There are a number of reasons to be skeptical because most of the evidence does not support Forbes' claim that the internet has killed inflation. 1. E-commerce affects only a small part of the Consumer Price Index As mentioned above, online shopping for goods represents 8.5% of total retail sales in the U.S. E-commerce is concentrated in four kinds of businesses (Table 1): Furniture & Home Furnishings (7% of total retail sales), Electronics & Appliances (20%), Health & Personal Care (15%), and Clothing (10%). Since goods make up 40% of the CPI, then 3.2% (8% times 40%) is a ballpark estimate for the size of goods e-commerce in the CPI. Table 1E-Commerce Market Share Of Goods Sector Did Amazon Kill The Phillips Curve? Did Amazon Kill The Phillips Curve? Table 2 shows the relative size of e-commerce in the service sector. The analysis is complicated by the fact that the data on services includes B-to-B sales in addition to B-to-C.2 However, e-commerce represents almost 4% of total sales for the service categories tracked by the BLS. Services make up 60% of the CPI, but the size drops to 26% if we exclude shelter (which is probably not affected by online shopping). Thus, e-commerce in the service sector likely affects 1% (3.9% times 26%) of the CPI. Table 2E-Commerce Market Share Of Service Sector Did Amazon Kill The Phillips Curve? Did Amazon Kill The Phillips Curve? Adding goods and services, online shopping affects about 4.2% of the CPI index at most. The bottom line is that the relatively small size of e-commerce at the consumer level limits any estimate of the impact of online sales on the broad inflation rate. 2. Most of the deceleration in inflation since 2007 has been in areas unaffected by e-commerce Table 3 compares the average contribution to annual average CPI inflation during 2000-2007 with that of 2007-2016. Average annual inflation fell from 2.9% in the seven years before the Great Recession to 1.8% after, for a total decline of just over 1 percentage point. The deceleration is almost fully explained by Energy, Food and Owners' Equivalent Rent. The bottom part of Table 3 highlights that the sectors with the greatest exposure to e-commerce had a negligible impact on the inflation slowdown. Table 3Comparison Of Pre- And Post-Lehman Inflation Rates Did Amazon Kill The Phillips Curve? Did Amazon Kill The Phillips Curve? 3. The cost advantages for online sellers are overstated Bain & Company, a U.S. consultancy, argues that e-commerce will not grow in importance indefinitely and come to dominate consumer spending.3 E-commerce sales are already slowing. Market share is following a classic S-shaped curve that, Bain estimates, will top out at under 30% by 2030. First, not everyone wants to buy everything online. Products that are well known to consumers and purchased on a regular basis are well suited to online shopping. But for many other products, consumers need to see and feel the product in person before making a purchase. Second, the cost savings of online selling versus traditional brick and mortar stores is not as great as many believe. Bain claims that many e-commerce businesses struggle to make a profit. The information technology, distribution centers, shipping, and returns processing required by e-commerce companies can cost as much as running physical stores in some cases. E-tailers often cannot ship directly from manufacturers to consumers; they need large and expensive fulfillment centers and a very generous returns policy. Moreover, online and offline sales models are becoming blurred. Retailers with physical stores are growing their e-commerce operations, while previously pure e-commerce plays are adding stores or negotiating space in other retailers' stores. Even Amazon now has storefronts. The shift toward an "multichannel" selling model underscores that there are benefits to traditional brick-and-mortar stores that will ensure that they will not completely disappear. 4. E-commerce is not the first revolution in the retail sector The retail sector has changed significantly over the decades and it is not clear that the disinflationary effect of the latest revolution, e-commerce, is any more intense than in the past. Economists at Goldman Sachs point out that the growth of Amazon's market share in recent years still lags that of Walmart and other "big box" stores in the 1990s (Chart 3).4 This fact suggests that "Amazonification" may not be as disinflationary as the previous big-box revolution. 5. Weak productivity growth and high profit margins are inconsistent with a large supply-side benefit from e-commerce As discussed above, economic theory suggests that a positive supply shock that cuts costs and boosts competition should trim profit margins and lift productivity. The problem is that the margins and productivity have moved in the opposite direction that economic theory would suggest (Chart 4). Chart 3Comparison Of Pre- And Post-Lehman Inflation Rates Did Amazon Kill The Phillips Curve? Did Amazon Kill The Phillips Curve? Chart 4Incompatible With A Supply Shock Incompatible With A Supply Shock Incompatible With A Supply Shock By definition, productivity rises when firms can produce the same output with fewer or cheaper inputs. However, it is well documented that productivity growth has been in a downtrend since the 1990s, and has been dismally low since the Great Recession. A Special Report from BCA's Global Investment Strategy 5 service makes a convincing case that mismeasurement is not behind the low productivity figures. In fact, in many industries it appears that productivity is over-estimated. If e-commerce is big enough to "move the dial" on overall inflation, it should be big enough to see in the aggregate productivity figures. Chart 5Retail Margin Squeeze Only In Department Stores Retail Margin Squeeze Only In Department Stores Retail Margin Squeeze Only In Department Stores One would also expect to see a margin squeeze across industries if e-commerce is indeed generating a lot of deflationary competitive pressure. Despite dismally depressed productivity, however, corporate profit margins are at the high end of the historical range across most of the sectors of the S&P 500. This is the case even in the retailing sector outside of department stores (Chart 5). These facts argue against the idea that the internet has moved the economy further toward a disinflationary "perfect competition" model. 6. Online price setting is characterized by frictions comparable to traditional retail We would expect to observe a low price dispersion across online vendors since the internet has apparently lowered the cost of monitoring competitors' prices and the cost of searching for the lowest price. We would also expect to see fairly synchronized price adjustments; if one vendor adjusts its price due to changing market conditions, then the rest should quickly follow to avoid suffering a massive loss of market share. However, a recent study of price-setting practices in the U.S. and U.K. found that this is not the case.6 The dataset covered a broad spectrum of consumer goods and sellers over a two-year period, comparing online with offline prices. The researchers found that market pricing "frictions" are surprisingly elevated in the online world. Price dispersion is high in absolute terms and on par with offline pricing. Academics for years have puzzled over high price rigidities and dispersion in retail stores in the context of an apparently stiff competitive environment, and it appears that online pricing is not much better. The study did not cover a long enough period to see if frictions were even worse in the past. Nonetheless, the evidence available suggests that the lower cost of monitoring prices afforded by the internet has not led to significant price convergence across sellers online or offline. Another study compared online and offline prices for multichannel retailers, using the massive database provided by the Billion Prices Project at MIT.7 The database covers prices across 10 countries. The study found that retailers charged the same price online as in-store in 72% of cases. The average discount was 4% for those cases in which there was a markdown online. If the observations with identical prices are included, the average online/offline price difference was just 1%. 7. Some measures of online prices have grown at about the same pace as the CPI index The U.S. Bureau of Labor Statistics does include online sales when constructing the Consumer Price Index. It even includes peer-to-peer sales by companies such as Airbnb and Uber. However, the BLS admits that its sample lags the popularity of such services by a few years. Moreover, while the BLS is trying to capture the rising proportion of sales done via e-commerce, "outlet bias" means that the CPI does not capture the price effect in cases where consumers are finding cheaper prices online. This is because the BLS weights the growth rate of online and offline prices, not the price levels. While there may be level differences, there is no reason to believe that the inflation rates for similar goods sold online and offline differ significantly. If the inflation rates are close, then the growing share of online sales will not affect overall inflation based on the BLS methodology. The BLS argues that any bias in the CPI due to outlet bias is mitigated to the extent that physical stores offer a higher level of service. Thus, price differences may not be that great after quality-adjustment. All this suggests that the actual consumer price inflation rate could be somewhat lower than the official rate. Nonetheless, it does not necessarily mean that inflation, properly measured, is being depressed by e-commerce to a meaningful extent. Indeed, Chart 6 highlights that the U.S. component of the Billion Prices Index rose at a faster pace than the overall CPI between 2009 and 2014. The Online Price Index fell in absolute and relative terms from 2014 to mid-2016, but rose sharply toward the end of 2016. Applying our guesstimate of the weight of e-commerce in the CPI (3.2% for goods), online price inflation added to overall annual CPI inflation by about 0.3 percentage points in 2016 (bottom panel of Chart 6). There is more deflation evident in the BLS' index of prices for Electronic Shopping and Mail Order Houses (Chart 7). Online prices fell relative to the overall CPI for most of the time since the early 1990s, with the relative price decline accelerating since the GFC. However, our estimate of the contribution to overall annual CPI inflation is only about -0.15 percentage points in June 2017, and has never been more than -0.3 percentage points. This could be an underestimate because it does not include the impact of services, although the service e-commerce share of the CPI is very small. Chart 6Online Price Index Online Price Index Online Price Index Chart 7Electronic Shopping Price Index Electronic Shopping Price Index Electronic Shopping Price Index Another way to approach this question is to focus on the parts of the CPI that are most exposed to e-commerce. It is impossible to separate the effect of e-commerce on inflation from other drivers of productivity. Nonetheless, if online shopping is having a significant deflationary impact on overall inflation, we should see large and persistent negative contributions from these parts of the CPI. We combined the components of the CPI that most closely matched the sectors that have high e-commerce exposure according to the BLS' annual Retail Survey (Chart 8). The sectors in our aggregate e-commerce price proxy include hotels/motels, taxicabs, books & magazines, clothing, computer hardware, drugs, health & beauty aids, electronics & appliances, alcoholic beverages, furniture & home furnishings, sporting goods, air transportation, travel arrangement and reservation services, educational services and other merchandise. The sectors are weighted based on their respective weights in the CPI. Our e-commerce price proxy has generally fallen relative to the overall CPI index since 2000. However, while the average contribution of these sectors to the overall annual CPI inflation rate has fallen in the post GFC period relative to the 2000-2007 period, the average difference is only 0.2 percentage points. The contribution has hovered around the zero mark for the past 2½ years. Surprisingly, price indexes have increased by more than the overall CPI since 2000 in some sectors where one would have expected to see significant relative price deflation, such as taxis, hotels, travel arrangement and even books. One could argue that significant measurement error must be a factor. How could the price of books have gone up faster than the CPI? Sectors displaying the most relative price declines are clothing, computers, electronics, furniture, sporting goods, air travel and other goods. We recalculated our e-commerce proxy using only these deflating sectors, but we boosted their weights such that the overall weight of the proxy in the CPI is kept the same as our full e-commerce proxy discussed above. In other words, this approach implicitly assumes that the excluded sectors (taxis, books, hotels and travel arrangement) actually deflated at the average pace of the sectors that remain in the index. Our adjusted e-commerce proxy suggests that online pricing reduced overall CPI inflation by about 0.1-to-0.2 percentage points in recent years (Chart 9). This contribution is below the long-term average of the series, but the drag was even greater several times in the past. Chart 8BCA E-Commerce Proxy Price Index BCA E-Commerce Proxy Price Index BCA E-Commerce Proxy Price Index Chart 9BCA E-Commerce Adjusted Proxy Price Index BCA E-Commerce Adjusted Proxy Price Index BCA E-Commerce Adjusted Proxy Price Index Admittedly, data limitations mean that all of the above estimates of the impact of e-commerce are ballpark figures. Conclusions We are keeping an open mind and reserving judgement on the disinflationary impact of robotics, artificial intelligence and the gig economy until we do more research. But in terms of the impact of e-commerce, it is difficult to find supportive evidence. The available data are admittedly far from ideal for confirming or disproving the "Amazonification" thesis. Perhaps better measures of e-commerce pricing will emerge in the future. Nonetheless, the measures available today do not suggest that online sales are depressing the overall inflation rate by more than 0.1 or 0.2 percentage points, and it does not appear that the disinflationary impact has intensified by much. One could argue that lower online prices are forcing traditional retailers to match the e-commerce vendors, allowing for a larger disinflationary effect than we estimate. Nonetheless, if this were the case, then we would expect to see significant margin compression in the retail sector. The sectors potentially affected by e-commerce make up a small part of the CPI index. The deceleration of inflation since the GFC has been in areas unaffected by online sales. High corporate profit margins and depressed productivity growth also argue against the idea that e-commerce represents a large positive macro supply shock. Finally, today's creative destruction in retail may be no more deflationary than the shift to 'big box' stores in the 1990s. Perhaps the main way that e-commerce is affecting the macro economy and financial markets is not through inflation, but via the reduction in the economy's capital spending requirement. Rising online activity means that we need fewer shopping malls and big box outlets to support a given level of consumer spending. This would reduce the equilibrium level of interest rates, since the Fed has to stimulate other parts of the economy to offset the loss of demand in capital spending in the retail sector. To the extent that central banks were slow to recognize that equilibrium rates had fallen to extremely low levels, then policy was behind the curve and this might have contributed to the current low inflation environment. Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com 1 Robert F. DeLucia, "Economic Perspective: A Nontraditional Analysis of Inflation," Prudential Capital Group (August 21, 2017). 2 Business to business, and business to consumer. 3 Aaron Cheris, Darrell Rigby and Suzanne Tager, "The Power Of Omnichannel Stores," Bain & Company Insights: Retail Holiday Newsletter 2016-2017 (December 19, 2016) 4 "US Daily: The Internet and Inflation: How Big is the Amazon Effect?" Goldman Sachs Economic Research (August 2, 2017). 5 Please see Global Investment Strategy Weekly Report, "Weak Productivity Growth: Don't Blame the Statisticians," dated March 25, 2016, available at gis.bcaresearch.com 6 Yuriy Gorodnichenko, Viacheslav Sheremirov, and Oleksandr Talavera, "Price Setting In Online Markets: Does IT Click?" Journal of the European Economic Association (July 2016). 7 Alberto Cavallo, "Are Online and Offline Prices Similar? Evidence from Large Multi-Channel Retailers," NBER Working Paper No. 22142 (March 2016).

A "culture of profound cost reduction" has gripped the business sector since the GFC according to one school of thought, permanently changing the relationship between labor market slack and wages or inflation. If true, it could mean that central banks are almost powerless to reach their inflation targets. Amazon, Airbnb, Uber, robotics, contract workers, artificial intelligence, horizontal drilling and driverless cars are just a few examples of companies and technologies that are cutting costs and depressing prices and wages. In the first of our series on inflation, we will focus on the rise of e-commerce and the related "Amazonification" of the economy. In theory, positive supply shocks should not have more than a temporary impact on inflation if the price level is indeed a monetary phenomenon in the long term. But a series of positive supply shocks could make it appear for quite a while that low inflation is structural in nature. We are keeping an open mind and reserving judgement on the disinflationary impact of robotics, artificial intelligence and the gig economy until we do more research. But in terms of the impact of e-commerce, it is difficult to find supportive evidence at the macro level. The admittedly inadequate measures of online prices available today do not suggest that e-commerce sales are depressing the overall inflation rate by more than 0.1 or 0.2 percentage points. Moreover, it does not appear that the disinflationary impact of competition in the retail sector has intensified over the years. Today's creative destruction in retail may be no more deflationary than the shift to 'big box' stores in the 1990s. Perhaps lower online prices are forcing traditional retailers to match the e-commerce vendors, allowing for a larger disinflationary effect than we estimate. However, the fact that retail margins are near secular highs outside of department stores argues against this thesis. The sectors potentially affected by e-commerce make up a small part of the CPI index. The deceleration of inflation since the GFC has been in areas unaffected by online sales. High profit margins for the overall corporate sector and depressed productivity growth also argue against the idea that e-commerce represents a large positive macro supply shock. Perhaps the main way that e-commerce is affecting the macro economy and financial markets is not through inflation, but via the reduction in the economy's capital spending requirement. This would reduce the equilibrium level of interest rates, since the Fed has to stimulate other parts of the economy to offset the loss of demand in capital spending in the retail sector. Anecdotal evidence is all around us. The global economy is evolving and it seems that all of the major changes are deflationary. Amazon, Airbnb, Uber, robotics, contract workers, artificial intelligence, horizontal drilling and driverless cars are just a few examples of companies and technologies that are cutting costs and depressing prices and wages. Central banks in the major advanced economies are having difficulty meeting their inflation targets, even in the U.S. where the labor market is tight by historical standards. Based on the depressed level of bond yields, it appears that the majority of investors believe that inflation headwinds will remain formidable for a long time. One school of thought is that low inflation reflects a lack of demand growth in the post-Great Financial Crisis (GFC) period. Another school points to the supply side of the economy. A recent report by Prudential Financial highlights "...obvious examples of ... new business models and new organizational structures, whereby higher-cost traditional methods of production, transportation, and distribution are displaced by more nontraditional cost-effective ways of conducting business."1 A "culture of profound cost reduction" has gripped the business sector since the GFC according to this school, permanently changing the relationship between labor market slack and wages or inflation (i.e., the Phillips Curve). Employees are less aggressive in their wage demands in a world where robots are threatening humans in a broadening array of industrial categories. Many feel lucky just to have a job. In a highly sensationalized article called "How The Internet Economy Killed Inflation," Forbes argued that "the internet has reduced many of the traditional barriers to entry that protect companies from competition and created a race to the bottom for prices in a number of categories." Forbes believes that new technologies are placing downward pressure on inflation by depressing wages, increasing productivity and encouraging competition. There are many factors that have the potential to weigh on prices, but analysts are mainly focusing on e-commerce, robotics, artificial intelligence, and the gig economy. In the first of our series on inflation, we will focus on the rise of e-commerce and the related "Amazonification" of the economy. The latter refers to the advent of new business models that cut out layers of middlemen between producers and consumers. Amazonification E-commerce has grown at a compound annual rate of more than 9% over the past 15 years, and now accounts for about 8½% of total U.S. retail sales (Chart II-1). Amazon has been leading the charge, accounting for 43% of all online sales in 2016 (Chart II-2). Amazon's business model not only cuts costs by eliminating middlemen and (until recently) avoiding expensive showrooms, but it also provides a platform for improved price discovery on an extremely broad array of goods. In 2013, Amazon carried 230 million items for sale in the United States, nearly 30 times the number sold by Walmart, one of the largest retailers in the world. Chart II-1E-Commerce: Steady Increase In Market Share E-Commerce: Steady Increase In Market Share E-Commerce: Steady Increase In Market Share Chart II-2Amazon Dominates September 2017 September 2017 With the use of a smartphone, consumers can check the price of an item on Amazon while shopping in a physical store. Studies show that it does not require a large price gap for shoppers to buy online rather than in-store. Amazon appears to be impacting other retailers' ability to pass though cost increases, leading to a rash of retail outlet closings. Sears alone announced the closure of 300 retail outlets this year. The devastation that Amazon inflicted on the book industry is well known. It is no wonder then, that Amazon's purchase of Whole Foods Market, a grocery chain, sent shivers down the spines of CEOs not only in the food industry, but in the broader retail industry as well. What would prevent Amazon from applying its model to furniture and appliances, electronics or drugstores? It seems that no retail space is safe. A Little Theory Before we turn to the evidence, let's review the macro theory related to positive supply shocks. The internet could be lowering prices by moving product markets toward the "perfect competition" model. The internet trims search costs, improves price transparency and reduces barriers to entry. The internet also allows for shorter supply chains, as layers of wholesalers and other intermediaries are removed and e-commerce companies allow more direct contact between consumers and producers. Fewer inventories and a smaller "brick and mortar" infrastructure take additional costs out of the system. Economic theory suggests that the result of this positive supply shock will be greater product market competition, increased productivity and reduced profitability. In the long run, workers should benefit from the productivity boost via real wage gains (even if nominal wage growth is lackluster). Workers may lower their reservation wage if they feel that increased competitive pressures or technology threaten their jobs. The internet is also likely to improve job matching between the unemployed and available vacancies, which should lead to a fall in the full-employment level of unemployment (NAIRU). Nonetheless, the internet should not have a permanent impact on inflation. The lower level of NAIRU and the direct effects of the internet on consumer prices discussed above allow inflation to fall below the central bank's target. The bank responds by lowering interest rates, stimulating demand and thereby driving unemployment down to the new lower level of NAIRU. Over time, inflation will drift back up toward target. In other words, a greater degree of the competition should boost the supply side of the economy and lower NAIRU, but it should not result in a permanently lower rate of inflation if inflation is indeed a monetary phenomenon and central banks strive to meet their targets. Still, one could imagine a series of supply shocks that are spread out over time, with each having a temporary negative impact on prices such that it appears for a while that inflation has been permanently depressed. This could be an accurate description of the current situation in the U.S. and some of the other major countries. We have sympathy for the view that the internet and new business models are increasing competition, cutting costs and thereby limiting price increases in some areas. But is there any hard evidence? Is the competitive effect that large, and is it any more intense than in the past? There are a number of reasons to be skeptical because most of the evidence does not support Forbes' claim that the internet has killed inflation. (1) E-commerce affects only a small part of the Consumer Price Index As mentioned above, online shopping for goods represents 8.5% of total retail sales in the U.S. E-commerce is concentrated in four kinds of businesses (Table II-1): Furniture & Home Furnishings (7% of total retail sales), Electronics & Appliances (20%), Health & Personal Care (15%), and Clothing (10%). Since goods make up 40% of the CPI, then 3.2% (8% times 40%) is a ballpark estimate for the size of goods e-commerce in the CPI. Table II-1E-Commerce Market Share Of Goods Sector (2015) September 2017 September 2017 Table II-2 shows the relative size of e-commerce in the service sector. The analysis is complicated by the fact that the data on services includes B-to-B sales in addition to B-to-C.2 However, e-commerce represents almost 4% of total sales for the service categories tracked by the BLS. Services make up 60% of the CPI, but the size drops to 26% if we exclude shelter (which is probably not affected by online shopping). Thus, e-commerce in the service sector likely affects 1% (3.9% times 26%) of the CPI. Table II-2E-Commerce Market Share Of Service Sector (2015) September 2017 September 2017 Adding goods and services, online shopping affects about 4.2% of the CPI index at most. The bottom line is that the relatively small size of e-commerce at the consumer level limits any estimate of the impact of online sales on the broad inflation rate. (2) Most of the deceleration in inflation since 2007 has been in areas unaffected by e-commerce Table II-3 compares the average contribution to annual average CPI inflation during 2000-2007 with that of 2007-2016. Average annual inflation fell from 2.9% in the seven years before the Great Recession to 1.8% after, for a total decline of just over 1 percentage point. The deceleration is almost fully explained by Energy, Food and Owners' Equivalent Rent. The bottom part of Table II-3 highlights that the sectors with the greatest exposure to e-commerce had a negligible impact on the inflation slowdown. Table II-3Comparison Of Pre- and Post-Lehman Inflation Rates September 2017 September 2017 (3) The cost advantages for online sellers are overstated Bain & Company, a U.S. consultancy, argues that e-commerce will not grow in importance indefinitely and come to dominate consumer spending.3 E-commerce sales are already slowing. Market share is following a classic S-shaped curve that, Bain estimates, will top out at under 30% by 2030. First, not everyone wants to buy everything online. Products that are well known to consumers and purchased on a regular basis are well suited to online shopping. But for many other products, consumers need to see and feel the product in person before making a purchase. Second, the cost savings of online selling versus traditional brick and mortar stores is not as great as many believe. Bain claims that many e-commerce businesses struggle to make a profit. The information technology, distribution centers, shipping, and returns processing required by e-commerce companies can cost as much as running physical stores in some cases. E-tailers often cannot ship directly from manufacturers to consumers; they need large and expensive fulfillment centers and a very generous returns policy. Moreover, online and offline sales models are becoming blurred. Retailers with physical stores are growing their e-commerce operations, while previously pure e-commerce plays are adding stores or negotiating space in other retailers' stores. Even Amazon now has storefronts. The shift toward an "multichannel" selling model underscores that there are benefits to traditional brick-and-mortar stores that will ensure that they will not completely disappear. (4) E-commerce is not the first revolution in the retail sector The retail sector has changed significantly over the decades and it is not clear that the disinflationary effect of the latest revolution, e-commerce, is any more intense than in the past. Economists at Goldman Sachs point out that the growth of Amazon's market share in recent years still lags that of Walmart and other "big box" stores in the 1990s (Chart II-3).4 This fact suggests that "Amazonification" may not be as disinflationary as the previous big-box revolution. (5) Weak productivity growth and high profit margins are inconsistent with a large supply-side benefit from e-commerce As discussed above, economic theory suggests that a positive supply shock that cuts costs and boosts competition should trim profit margins and lift productivity. The problem is that the margins and productivity have moved in the opposite direction that economic theory would suggest (Chart II-4). Chart II-3Amazon Vs. Walmart: ##br##Who's More Deflationary? September 2017 September 2017 Chart II-4Incompatible With A Supply Shock Incompatible With A Supply Shock Incompatible With A Supply Shock By definition, productivity rises when firms can produce the same output with fewer or cheaper inputs. However, it is well documented that productivity growth has been in a downtrend since the 1990s, and has been dismally low since the Great Recession. A Special Report from BCA's Global Investment Strategy5 service makes a convincing case that mismeasurement is not behind the low productivity figures. In fact, in many industries it appears that productivity is over-estimated. If e-commerce is big enough to "move the dial" on overall inflation, it should be big enough to see in the aggregate productivity figures. Chart II-5Retail Margin Squeeze ##br##Only In Department Stores Retail Margin Squeeze Only In Department Stores Retail Margin Squeeze Only In Department Stores One would also expect to see a margin squeeze across industries if e-commerce is indeed generating a lot of deflationary competitive pressure. Despite dismally depressed productivity, however, corporate profit margins are at the high end of the historical range across most of the sectors of the S&P 500. This is the case even in the retailing sector outside of department stores (Chart II-5). These facts argue against the idea that the internet has moved the economy further toward a disinflationary "perfect competition" model. (6) Online price setting is characterized by frictions comparable to traditional retail We would expect to observe a low price dispersion across online vendors since the internet has apparently lowered the cost of monitoring competitors' prices and the cost of searching for the lowest price. We would also expect to see fairly synchronized price adjustments; if one vendor adjusts its price due to changing market conditions, then the rest should quickly follow to avoid suffering a massive loss of market share. However, a recent study of price-setting practices in the U.S. and U.K. found that this is not the case.6 The dataset covered a broad spectrum of consumer goods and sellers over a two-year period, comparing online with offline prices. The researchers found that market pricing "frictions" are surprisingly elevated in the online world. Price dispersion is high in absolute terms and on par with offline pricing. Academics for years have puzzled over high price rigidities and dispersion in retail stores in the context of an apparently stiff competitive environment, and it appears that online pricing is not much better. The study did not cover a long enough period to see if frictions were even worse in the past. Nonetheless, the evidence available suggests that the lower cost of monitoring prices afforded by the internet has not led to significant price convergence across sellers online or offline. Another study compared online and offline prices for multichannel retailers, using the massive database provided by the Billion Prices Project at MIT.7 The database covers prices across 10 countries. The study found that retailers charged the same price online as in-store in 72% of cases. The average discount was 4% for those cases in which there was a markdown online. If the observations with identical prices are included, the average online/offline price difference was just 1%. (7) Some measures of online prices have grown at about the same pace as the CPI index The U.S. Bureau of Labor Statistics does include online sales when constructing the Consumer Price Index. It even includes peer-to-peer sales by companies such as Airbnb and Uber. However, the BLS admits that its sample lags the popularity of such services by a few years. Moreover, while the BLS is trying to capture the rising proportion of sales done via e-commerce, "outlet bias" means that the CPI does not capture the price effect in cases where consumers are finding cheaper prices online. This is because the BLS weights the growth rate of online and offline prices, not the price levels. While there may be level differences, there is no reason to believe that the inflation rates for similar goods sold online and offline differ significantly. If the inflation rates are close, then the growing share of online sales will not affect overall inflation based on the BLS methodology. The BLS argues that any bias in the CPI due to outlet bias is mitigated to the extent that physical stores offer a higher level of service. Thus, price differences may not be that great after quality-adjustment. All this suggests that the actual consumer price inflation rate could be somewhat lower than the official rate. Nonetheless, it does not necessarily mean that inflation, properly measured, is being depressed by e-commerce to a meaningful extent. Indeed, Chart II-6 highlights that the U.S. component of the Billion Prices Index rose at a faster pace than the overall CPI between 2009 and 2014. The Online Price Index fell in absolute and relative terms from 2014 to mid-2016, but rose sharply toward the end of 2016. Applying our guesstimate of the weight of e-commerce in the CPI (3.2% for goods), online price inflation added to overall annual CPI inflation by about 0.3 percentage points in 2016 (bottom panel of Chart II-6). There is more deflation evident in the BLS' index of prices for Electronic Shopping and Mail Order Houses (Chart II-7). Online prices fell relative to the overall CPI for most of the time since the early 1990s, with the relative price decline accelerating since the GFC. However, our estimate of the contribution to overall annual CPI inflation is only about -0.15 percentage points in June 2017, and has never been more than -0.3 percentage points. This could be an underestimate because it does not include the impact of services, although the service e-commerce share of the CPI is very small. Chart II-6Online Price Index Online Price Index Online Price Index Chart II-7Electronic Shopping Price Index Electronic Shopping Price Index Electronic Shopping Price Index Another way to approach this question is to focus on the parts of the CPI that are most exposed to e-commerce. It is impossible to separate the effect of e-commerce on inflation from other drivers of productivity. Nonetheless, if online shopping is having a significant deflationary impact on overall inflation, we should see large and persistent negative contributions from these parts of the CPI. We combined the components of the CPI that most closely matched the sectors that have high e-commerce exposure according to the BLS' annual Retail Survey (Chart II-8). The sectors in our aggregate e-commerce price proxy include hotels/motels, taxicabs, books & magazines, clothing, computer hardware, drugs, health & beauty aids, electronics & appliances, alcoholic beverages, furniture & home furnishings, sporting goods, air transportation, travel arrangement and reservation services, educational services and other merchandise. The sectors are weighted based on their respective weights in the CPI. Our e-commerce price proxy has generally fallen relative to the overall CPI index since 2000. However, while the average contribution of these sectors to the overall annual CPI inflation rate has fallen in the post GFC period relative to the 2000-2007 period, the average difference is only 0.2 percentage points. The contribution has hovered around the zero mark for the past 2½ years. Surprisingly, price indexes have increased by more than the overall CPI since 2000 in some sectors where one would have expected to see significant relative price deflation, such as taxis, hotels, travel arrangement and even books. One could argue that significant measurement error must be a factor. How could the price of books have gone up faster than the CPI? Sectors displaying the most relative price declines are clothing, computers, electronics, furniture, sporting goods, air travel and other goods. We recalculated our e-commerce proxy using only these deflating sectors, but we boosted their weights such that the overall weight of the proxy in the CPI is kept the same as our full e-commerce proxy discussed above. In other words, this approach implicitly assumes that the excluded sectors (taxis, books, hotels and travel arrangement) actually deflated at the average pace of the sectors that remain in the index. Our adjusted e-commerce proxy suggests that online pricing reduced overall CPI inflation by about 0.1-to-0.2 percentage points in recent years (Chart II-9). This contribution is below the long-term average of the series, but the drag was even greater several times in the past. Chart II-8BCA E-Commerce Proxy Price Index BCA E-Commerce Proxy Price Index BCA E-Commerce Proxy Price Index Chart II-9BCA E-Commerce Adjusted Proxy Price Index BCA E-Commerce Adjusted Proxy Price Index BCA E-Commerce Adjusted Proxy Price Index Admittedly, data limitations mean that all of the above estimates of the impact of e-commerce are ballpark figures. Conclusions We are keeping an open mind and reserving judgement on the disinflationary impact of robotics, artificial intelligence and the gig economy until we do more research. But in terms of the impact of e-commerce, it is difficult to find supportive evidence. The available data are admittedly far from ideal for confirming or disproving the "Amazonification" thesis. Perhaps better measures of e-commerce pricing will emerge in the future. Nonetheless, the measures available today do not suggest that online sales are depressing the overall inflation rate by more than 0.1 or 0.2 percentage points, and it does not appear that the disinflationary impact has intensified by much. One could argue that lower online prices are forcing traditional retailers to match the e-commerce vendors, allowing for a larger disinflationary effect than we estimate. Nonetheless, if this were the case, then we would expect to see significant margin compression in the retail sector. The sectors potentially affected by e-commerce make up a small part of the CPI index. The deceleration of inflation since the GFC has been in areas unaffected by online sales. High corporate profit margins and depressed productivity growth also argue against the idea that e-commerce represents a large positive macro supply shock. Finally, today's creative destruction in retail may be no more deflationary than the shift to 'big box' stores in the 1990s. Perhaps the main way that e-commerce is affecting the macro economy and financial markets is not through inflation, but via the reduction in the economy's capital spending requirement. Rising online activity means that we need fewer shopping malls and big box outlets to support a given level of consumer spending. This would reduce the equilibrium level of interest rates, since the Fed has to stimulate other parts of the economy to offset the loss of demand in capital spending in the retail sector. To the extent that central banks were slow to recognize that equilibrium rates had fallen to extremely low levels, then policy was behind the curve and this might have contributed to the current low inflation environment. Mark McClellan Senior Vice President The Bank Credit Analyst 1 Robert F. DeLucia, "Economic Perspective: A Nontraditional Analysis Of Inflation," Prudential Capital Group (August 21, 2017). 2 Business to business, and business to consumer. 3 Aaron Cheris, Darrell Rigby and Suzanne Tager, "The Power Of Omnichannel Stores," Bain & Company Insights: Retail Holiday Newsletter 2016-2017 (December 19, 2016). 4 "US Daily: The Internet And Inflation: How Big Is The Amazon Effect?" Goldman Sachs Economic Research (August 2, 2017). 5 Please see Global Investment Strategy Weekly Report, "Weak Productivity Growth: Don't Blame The Statisticians," dated March 25, 2016, available at gis.bcaresearch.com 6 Yuriy Gorodnichenko, Viacheslav Sheremirov, and Oleksandr Talavera, "Price Setting In Online Markets: Does IT Click?" Journal of the European Economic Association (July 2016). 7 Alberto Cavallo, "Are Online And Offline Prices Similar? Evidence From Large Multi-Channel Retailers," NBER Working Paper No. 22142 (March 2016).