Financial Markets

Special Report

Highlights We are upgrading our allocation to Indian stocks from neutral to overweight within EM equity portfolios. India's public banks are much further along in their necessary adjustment process, and the credit cycle downturn is much more advanced relative to China's. To capitalize on this theme, we recommend going long Indian banks and shorting Chinese bank stocks. India's public bank recapitalization program will allow them to slowly augment credit origination, assisting the economic recovery. Feature Chart I-1Favor Indian Banks Versus Chinese Ones Our report this week highlights the results from stress tests we conducted on Indian and Chinese public banks, and also compares their respective equity valuations. Based on our findings, we are initiating a new relative equity trade: long Indian / short Chinese bank stocks (Chart I-1). The health of the banking system, the credit cycle outlook as well as the performance of bank share prices hold the key to relative performance of any bourse in the EM universe. Provided our positive bias toward Indian banks relative to their EM peers on all the above parameters, we are upgrading our allocation to India from neutral to overweight within EM equity portfolios. Indian Versus Chinese Public Banks From 2003 to 2012, India went through a large credit binge and capital misallocation cycle in its industrial and infrastructure sectors. During this period, banks' loans to companies and bank assets rose from 12% to 23% and 63% to 85% of GDP, respectively (Chart I-2A). By comparison, Chinese (ex-policy) commercial banks' claims on companies and their total assets have surged from 85% to 110% and from under 180% to 230% of GDP, respectively, since 2009 (Chart I-2B). In both countries, the banking sector remains dominated by public banks that hold more than 50% of banking system assets. Chart I-2ACredit Boom In Perspective: India Chart I-2BCredit Boom In Perspective: China Today, Indian public banks - who were the main lenders to industrial companies during the corporate credit binge in the 2003-12 period - have been experiencing mushrooming bad loans. Total public banks' NPLs and distressed asset ratios have reached 13.5% and 2.7% of total loans, respectively (Chart I-3). By contrast, for all Chinese banks, the current NPL ratio is at a mere 1.7%, while the distressed loan ratio stands at only 3.6% of total loans. Chart I-3NPL Ratios In Perspective: India & China Further, under pressure from the central bank, Indian public banks have been raising provisioning levels for bad assets very aggressively. On the flip side, Chinese regulators have been following tolerant policies toward their own commercial banks. As such, the provisions-to-loans ratio at all public banks now stands at 3% in China, compared with 5.6% in India. In addition, Chinese banks have bought a lot of corporate bonds that are not provisioned for at all. Does this higher NPL ratio in India relative to China mean that credit allocation is much worse in India? Not quite. The thesis that Indian public banks are more poorly managed than Chinese public banks is not accurate. These banks are managed by public sector executives who often allocate credit to support government growth policies. This is why it is reasonable to assume that the quality of credit allocation among Chinese and Indian public banks is probably similar. As such, we presume that Chinese banks' current NPL ratio is severely understated, and has the potential to rise to levels currently being reported by Indian public banks. The basis is that the Chinese credit boom has dramatically exceeded that of India (see Chart I-2A and I-2B on page 2). Typically, the resulting NPL ratio is proportional to the magnitude of the preceding credit frenzy. Finally, India's central government announced a major recapitalization plan in October 2017 to assist the country's public banks in cleaning up their balance sheets and to also support them in expanding credit. It is likely, therefore, that these banks are now approaching the final stages of their balance sheet repair and deleveraging process. Bottom Line: India's public banks are much further along in their necessary adjustment, and their credit cycle downturn is also much more advanced relative to Chinese banks. The latter have been postponing the inevitable balance sheet clean-up process. To capitalize on this theme, we recommend going long Indian banks and shorting Chinese bank stocks. Banking Stress Test For India And China We have conducted stress tests for India's top seven and China's top five listed public banks. We used the following assumptions for the three scenarios we considered: Non-performing risk-weighted assets (NPA) ratios to rise to 14% (pessimistic), 12% (baseline) and 10% (optimistic scenario) of risk-weighted assets for both Indian and Chinese public banks. Risk-weighted assets adjust banks' various types of assets based on their degree of riskiness. In that way, the risk-weighted asset values are comparable between the two banking systems. We assume a 30% recovery rate in all three NPA scenarios for both countries. The recovery rate on Chinese banks' NPAs in the 2001-2005 period was 20% amid a booming economy. The assumed recovery rate of 30% is therefore not low. The outcome of the stress tests is as follows: In the baseline scenario of 12% NPA, the losses post recovery and provisions would amount to 1.3 trillion rupees in India (0.9% of GDP) and RMB 3.4 trillion in China (4.2% of GDP). This would translate into a 33% equity impairment for India's seven public banks, and 48% for China's five public banks (Table I-1 and I-2, column 7). Table I-1Stress Test For Top 7 Indian Public Banks Table I-2Stress Test For Top 5 Chinese Public Banks From a valuation standpoint, the post-impairment price-to-book value (PBV) ratio would jump to 1.44 and 1.62 for Indian- and Chinese-listed public banks, respectively. Assuming a fair PBV ratio of 1.3 - which is the average PBV ratio for all EM banks since 2011 - Indian public banks are 11% overvalued and Chinese ones are about 25% overvalued. In other words, if one were to calculate the true PBV ratio of these banks after a comprehensive "clean-up" has been done, then Indian public bank stocks would be cheaper than Chinese ones. It is important to note that the above valuation exercise does not take into consideration banks' future profits. As such, we account for their recurring profits in the following manner: Table I-3 calculates the ratio of NPA losses to banks' recurring net profits before provisioning. Losses are the amount to be written-off post provisioning and recovery. In the baseline scenario of a 12% of NPA, this ratio is 2.5 for India and 3.4 for China. In other words, it will take 2.5 and 3.4 years of net profits before provisions close the "black hole" of NPA losses (post provisions and recovery) in India and China, respectively. Hence, on this measure as well, India's listed public banks appear more appealing than those in China. Table I-3Profit Coverage Of Loan Losses There is a caveat regarding Chinese banks' stress and their post-impairment book value. Our analysis is performed based on risk-weighted assets, and does not include off-balance-sheet assets. Therefore, any losses from off-balance-sheet assets will make losses for Chinese public banks greater than our analysis captures. Further, the Chinese financial authorities are currently tightening regulations, which will likely curtail banks' off-balance-sheet activities and by extension their profitability. These risks are not present in India, where banks have less off-balance-sheet assets. Bottom Line: Public bank stocks are currently overvalued by about 11% and 25% in absolute terms in both India and China, respectively. This favors Indian bank share prices outperforming their Chinese peers. The fact that the "clean-up" has not yet begun in China reinforces this trade. Banks' Recapitalization In India Saddled with NPLs, Indian public banks have not been willing to lend in recent years. Chart I-4 demonstrates that their loan growth has stalled. Credit to large industrial companies has in particular suffered (Chart I-4, bottom panel), as most of this type of credit is typically extended by public banks. Chart I-4India: Public Bank Loan Growth Has Slumped Consequently, India's capital expenditures have languished in recent years, weighing not only on cyclical growth but also depressing long-term productivity and potential growth. In October, the Indian government announced an estimated 2.11 trillion rupees public bank recapitalization program that will be implemented over the next two years. The program is for all public banks, while the above stress test was performed for only the top seven listed public banks. The latter account for around 60% of all public banks' assets, so we assume they will get around 60% of the stated recapitalization amount. The recapitalization program is designed as follows: The central government plans to inject 180 billion rupees of equity capital into all public banks via budgetary allocations. The public banks will in turn raise 580 billion rupees from the market. The remaining 1,350 billion rupees will come from government-issued Bank Recapitalization Bonds. The government will issue bonds to banks and then use the funds to buy more shares from public banks. It is important to note that in the stress test above and for the calculation of post-impairment PBV ratios, we assume the government will not subsidize existing shareholders when it injects money into public banks. This means the government will provide equity capital to public banks at post-impairment equity value - i.e., at a fair market price. It will be difficult for the Indian government to bail out its public banks without making current shareholders bear losses. If the government bails out public banks' private and foreign shareholders, the opposition parties will use the bank recapitalization program against Prime Minister Narendra Modi's government in the general elections scheduled to be held in 2019. Many investors and commentators assume that India's bank recapitalization program is automatically bullish for bank share prices. While it is positive for banks' ability to lend and drive growth in the medium and long term, the program is not necessarily bullish for share prices, particularly at their current high levels. The same is true for potential recapitalization programs in China. Overall, odds are that current shareholders of public banks will likely shoulder meaningful losses in India and possibly in China as well. How well off will capitalized public banks in India be after implementation of the recapitalization program? In the case of the seven Indian public banks we performed the stress test on, Table I-4 estimates that post-impairment and recovery, the total equity capital-to-risk-weighted assets ratio will be 8% in our baseline scenario. This is lower than the regulatory minimum of 9%. Table I-4Capital Ratios For India's Top 7 Public Banks The recapitalization will bring this equity capital adequacy ratio to 11.3%, which exceeds the regulatory minimum of 9%. Hence, after the program is completed, Indian public banks will likely become well capitalized and will be able to resume their lending and expand their assets. This in turn will facilitate the economic recovery. Bottom Line: The Indian government's recapitalization program is sufficient to raise public banks' capital adequacy ratio above the regulatory minimum. This will allow public banks to resume their lending. India's Cyclical Growth Outlook India's cyclical outlook will be one of muted recovery. Yet it is superior to other EMs, where we expect meaningful deceleration due to a potential slowdown in China and a rollover in commodities prices. Public banks' recap program will be slow in India - to be conducted over the next two years - and banks' ability to boost lending will improve only gradually. Meanwhile, private banks have and will probably continue to concentrate their lending efforts on consumers rather than on industrial companies and infrastructure. In the next 12-18 months, a slow improvement in public banks' ability to originate credit will allow only moderate improvement in capital spending growth. The latter is required to resolve bottlenecks and unleash the nation's productivity potential. Several indicators of capital spending are lukewarm (Chart I-5, top panel). However, new capex project announcements and the number of investment proposals have been dropping (Chart I-5, middle panel). Surprisingly, companies' foreign external borrowing is still contracting, despite booming capital inflows into EM (Chart I-5, bottom panel). On the consumer side, the outlook remains bright. Motorcycle sales have recovered sharply and commercial vehicle sales are beginning to pick up (Chart I-6). Chart I-5India's Capital Spending Is Sluggish Chart I-6Indian Consumer Health Is Strong Consumer/personal loans are accelerating from an already strong growth rate, largely thanks to the aggressiveness of private sector banks (Chart I-6, bottom panel). In turn, the employment outlook is finally beginning to show signs of improvement (Chart I-7). The manufacturing PMI has also risen substantially, and is currently in expansion territory (Chart I-8). Likewise, the service sector PMI has bounced above 50. Chart I-7India's Employment Is Turning The Corner Chart I-8India: PMIs Are Positive Finally, India is less exposed to China's growth and a retracement in commodities prices than many other emerging economies. This makes us upbeat on India's cyclical economic dynamics and relative equity and currency performance versus other EMs. Bottom Line: India's cyclical outlook is better than that of many other EMs. Structural Tailwinds And Impediments India holds huge promise for investors as it is a much-underinvested economy, and potential return on capital is considerably higher in those countries than in relatively overinvested ones. In addition, its population and labor force growth are among the highest in mainstream developing countries. On the other hand, for such potential to be realized, the country needs to be able to boost its productivity. On this count, the outlook is less positive. India's share of global goods and services exports has declined substantially since 2011 (Chart I-9). This should not be surprising, given weak investment spending has led to stagnation in trade competitiveness. Chart I-10 reveals that based on the UNCTAD1 dataset, India has been losing market share in both low- and high-skilled labor sectors export markets worldwide. Chart I-9India's Share In Global Trade Chart I-10India Has Been Losing Export Market Share While certain reforms such as the introduction of a sales tax will have a positive impact on the economy, other much-needed changes, such as land and labor market reforms, have so far remained unattainable. Moreover, the agriculture sector still faces material challenges. Without these vital reforms, it will be difficult to boost efficiency and productivity and build global competitiveness. Finally, in terms of education enrollment, India lags other EMs, especially China, in tertiary education (Chart I-11). This makes it even more difficult to boost productivity and growth potential. Bottom Line: India has great secular potential, but the structural advance has stalled since 2011. The jury is still out on whether it can implement additional reforms to realize this potential. Investment Conclusions India's banking sector outlook is brighter, and the deleveraging cycle is much more advanced, compared with many other EMs in general and China in particular. Therefore, we recommend a new relative equity trade: long Indian banks / short Chinese banks. Investors could buy Indian public banks or all banks with the understanding that private banks are typically in better shape than their state-owned peers, but are also much more expensive. We will be tracking this trade's performance using the Bankex index for India and the MSCI bank index for China. The Bankex index has a larger share of market cap of public banks than the MSCI India bank index. Within China, we are maintaining our short small and medium / long large banks position initiated on October 26th 2016. We are also recommending EM equity investors upgrade the Indian bourse from neutral to overweight. We shifted Indian stocks from overweight to neutral on August 23rd 2017, but the risk-reward has improved since then (Chart I-12). Chart I-11India's Education Improvement Is Lagging Chart I-12Upgrade Indian Bourse Within EM Universe Our primary concerns with EM stocks are a China slowdown, a rollover in commodities prices and a rebound in the U.S. dollar. Associated strains in countries with large foreign debt levels or wide current account deficits as well as lack of credit deleveraging and bank recapitalization will define EM financial markets' performance in the next 12-18 months. On all of these counts, India scores better than many EMs, justifying this equity upgrade. The absolute outlook for Indian stocks, however, is not inspiring. This equity market is rather expensive and overbought in absolute terms. If EM risk assets experience a setback in 2018, as we expect, Indian equities will also relapse in absolute terms. Ayman Kawtharani, Associate Editor ayman@bcaresearch.com Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com   1 United Nations Conference on Trade and Development.   Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Special Report

Equities have melted up in recent weeks, celebrating the tax bill passage, synchronized upswing in global economic data, still quiescent inflation and near vanishing tail risk. On July 10th when we penned the "SPX 3,000?" report, the S&P 500 was close to 2400.1 Over the past six months stocks have been in an uninterrupted upleg, moving to within 10% of our SPX 3,000 target. Table 1 Stocks have run "too far too fast" for our liking and there are increasing odds of a healthy pullback, especially now that no pundits are talking of a correction. In addition, were the selloff in the bond markets to accelerate in a short time frame, at some point it will cause equity market consternation. But, bonds still remain extremely overvalued versus stocks (Chart 1). Late last year, we began to modestly de-risk the portfolio via booking impressive gains in tactical market-neutral trades, as our upbeat cyclical view remains intact.2 Our cyclical strategy is to "buy the dip", as we do not foresee a recession in the coming 9-12 months. Importantly, profits will dictate the S&P 500's direction and the cyclical path of least resistance is higher still. Our SPX profit model continues to forecast healthy EPS growth in 2018 (Chart 2) and as we posited in the last report of 2017, earnings will do the heavy lifting at the current juncture with the forward P/E multiple likely moving laterally (Chart 3). Chart 1Simple Bond Valuation Metric Says:##br## Bonds Are Overvalued Vs. Stocks Chart 2All ##br##Clear Chart 3EPS Will Do The##br## Heavy Lifting In 2018 A simple decomposition shows that equity returns could reasonably reach a low-to-mid double digit level this year. Our assumptions are the following: nominal GDP can grow near 5% (3% real plus 2% inflation) and thus we estimate organic EPS growth that typically mimics GDP at this stage of the cycle of ~5%, ~2% dividend yield, ~2% buyback yield, ~5% tax related boost to EPS and no multiple expansion. The above assumptions are based on four key drivers: energy and financials will command a larger slice of the earnings pie,3 synchronized global capex upcycle will boost EPS,4 delayed positive translation effects from the U.S. dollar will lift profits5 and easy fiscal policy will also act as a tonic to EPS.6 On this note, this White Paper officially introduces the U.S. Equity Strategy earnings models for the eleven GICS1 equity sectors. We have identified key macro earnings drivers for each sector and incorporated them into individual sector models. The objective is to forecast the direction of earnings growth. Beyond introducing our EPS models, the purpose of this White Paper is to also compare and contrast the cyclical readings of our equity sector models with sell-side analysts' profit growth (Charts 4 & 5) and margin expectations and help clients position portfolios for the rest of 2018. The earnings models carry the most weight in determining our sector positioning, with our macro overlay and our valuation and technical indicators rounding out our methodology. Currently, our earnings models are consistent with maintaining a mostly cyclically biased portfolio structure (top panel, Chart 6), and thus participating in the broad market's overshoot. Chart 4What EPS Are Priced In... Chart 5...Per Sector For 2018 Chart 6Continue To Prefer Cyclicals Over Defensives Encouragingly, an equal weight of the 10 GICS1 sector model outputs (we are excluding real estate due to lack of history), accurately forecasts the S&P 500's profit growth (bottom panel, Chart 6), and currently also confirms the broad market's upbeat four factor macro EPS model (Chart 2). Anastasios Avgeriou, Vice President U.S. Equity Strategy anastasios@bcaresearch.com Financials (Overweight) Our financials earnings growth model comprises bank credit growth, the U.S. dollar index and net earnings revisions. The U.S. credit impulse is gaining traction, indicating that the market has digested the almost doubling in long-term rates over the past 18 months. Bankers are willing extenders of C&I credit and, with the economy humming north of 3% in real GDP terms, the outlook for loan growth is excellent. Loosening U.S. banking regulatory requirements, and pent up demand for shareholder friendly activities are all welcome news for financials profitability. Tack on BCA's higher interest rate view in 2018 and net interest margins will also get a bump, further adding to the sector's EPS euphoria. Credit quality is the third key profit driver for bank profitability and pristine credit quality is a harbinger of increased profits. The unemployment rate is plumbing generational lows and suggests that non-performing loans as a percentage of total loans will remain on a downward trajectory. Our profit model is expanding at twice the current profit growth rate (second panel, Chart 7) and 10 percentage points above the Street's 12-month forward estimates (top panel, Chart 5). In fact, the latter have gone vertical of late playing catch up to our model's estimates. The S&P financials sector remains a core portfolio overweight and we reiterate our high-conviction overweight status in the heavyweight S&P banks index. Chart 7Financials (Overweight) Energy (Overweight) The three drivers behind the S&P energy sector EPS growth model are oil-related currencies, the U.S. oil & gas rig count and WTI crude oil prices. A depreciating greenback, whittling down OECD oil stocks and rising global oil demand are all boosting energy profitability. OPEC 2.0 cutbacks have not only helped stabilize oil markets, but also paved the way for a breakout in oil prices above the $62.50/bbl stiff resistance level. Sustained OPEC output restraint will counterbalance U.S. shale oil production increases and coupled with rising global demand likely continue to underpin oil prices. Our synchronized global capex upcycle theme included the basic resources following a multi-year drubbing in outlays. Energy capex cannot contract at double digit rates indefinitely. Already a V-shaped capex momentum recovery is in store, as 2018 capital spending budgets are on track to at least match 2017. Our EPS growth model (second panel, Chart 8) matches sell-side analyst optimism (third panel, Chart 5). Keep in mind that only recently did the energy space become profit positive, making a solid recovery from an extremely low base. Margins are only now renormalizing above the zero line and breakneck pace EPS growth should continue in 2018. Following a negative 2017 return, the S&P energy sector is the best performing sector year-to-date, and we reiterate the high-conviction overweight stance. Chart 8Energy (Overweight) Industrials (Overweight) Our S&P industrials EPS model comprises the ISM manufacturing survey, raw industrials commodity prices and interest rates. It has an excellent track record in forecasting industrials EPS momentum, and sports one of the highest explanatory powers amongst all sector EPS models. While industrials EPS growth has been bouncing off the zero line for the better part of the past five years, our profit model has spoken: forecast EPS are in a V-shaped recovery since the end of the recent manufacturing recession (second panel, Chart 9). Commodity prices are recovering and increasing final demand, coupled with a soft U.S. dollar suggest that more gains are in store. Tack on the global virtuous capex upcycle, and the stars are aligned for this deep cyclical sector to break out of its multi-year trading range funk on the back of a surge in profits. China is a wild card, but signs of stability are enough to sustain the upward trajectory in the commodity-levered complex, including industrials stocks. Our industrials sector EPS model suggests that industrials profits will easily surpass the low (and below the overall market) analysts' EPS growth hurdle (third panel, Chart 4). The late-cyclical S&P industrials sector remains an overweight. Chart 9Industrials (Overweight) Consumer Staples (Overweight) The S&P consumer staples EPS growth model key drivers are: food exports, non-discretionary retail sales and analysts' net earnings revision ratio. Overall industry exports are expanding at a healthy clip as a consequence of a softening U.S. dollar and robust European and rebounding emerging markets demand. Deflating raw food commodity prices are offsetting rising energy and labor input costs, heralding a sideways move to margins. Sell side analysts are also currently penciling in a lateral profit margin move (middle panel, Chart 10). Our model is expanding at a near double digit rate, and is in line with 12-month forward EPS growth estimates (second panel, Chart 4). Investors have been vehemently avoiding staples stocks during the board market's uninterrupted run up, and have put out positioning offside. However, in the context of our cyclical over defensive portfolio bent we refrain from putting all our eggs in one basket, and prefer to keep consumer staples as our sole defensive sector overweight. This small hedge will serve our portfolio well if we do indeed get a healthy Q1/2018 pullback, as we expect. Chart 10Consumer Staples (Overweight) Consumer Discretionary (Neutral - Downgrade Alert) Measures of consumer confidence, consumer discretionary exports and the net earnings revisions ratio comprise BCA's global consumer discretionary EPS growth model, which has an excellent track record in forecasting the path of consumer discretionary profits. Consumer confidence is rolling over, albeit from a nose-bleed level, signaling that, at the margin, discretionary consumer outlays will remain tame. Worrisomely, rising interest rates coupled with a breakout in crude oil prices are net negatives for consumer spending. Our consumer drag indicator captures these consumer headwinds and warns that the sector is not out of the woods yet (bottom panel, Chart 11). The Fed is on track to raise rate three more times in 2018 and continue to mop up liquidity via renormalizing its balance sheet. This dual tightening backdrop bodes ill for early cyclical discretionary stocks as we highlighted in the September 25th Weekly Report. Our consumer discretionary EPS growth model is making an effort to bounce, signaling that contracting earnings will likely reverse course and come out of their recent funk (second panel). But, analysts are overly optimistic penciling in a near double-digit profit growth backdrop for the consumer discretionary sector (fourth panel, Chart 5). Netting it all out, the anemic message from our profit model along with the ongoing Fed tightening cycle and spiking energy prices warrant a downgrade alert. Stay tuned. Chart 11Consumer Discretionary (Neutral-Downgrade Alert) Telecom Services (Neutral) Telecom pricing power and capital expenditures expectations comprise our S&P telecom services EPS growth model. Telecom capital expenditures have bounced off the zero line and are growing at 4% per annum while sector sales growth has been nil. This capital-intensive industry must continually invest to stay relevant. A push by telecom carriers into TV offerings as part of a quad-play (internet, wireline, wireless and TV) has rekindled an M&A boom, and capex is slated to increase. However, margins will suffer if increased investment fails to translate into new sales (bottom panel, Chart 12). Steeply contracting pricing power is a bad omen both for top and bottom line growth prospects (fourth panel). Hopefully, industry consolidation will lead to a better pricing backdrop, but the jury is still out. Our EPS model has sunk into the contraction zone (second panel). Analysts are a little bit more sanguine, penciling in low single-digit profit growth (bottom panel, Chart 4). Industry deflation is not alone as a headwind as the bond market selloff is weighing on the high dividend yielding telecom services stocks. Despite all the bearish news, near all-time lows in relative valuation and washed out technicals are keeping us on the sidelines. Chart 12Telecom Services (Neutral) Materials (Neutral) Materials EPS growth is a far cry from the near 100% year-over-year mark hit during the commodity super-cycle the mid-2000s and the reflex rebound following the Great Recession (second panel, Chart 13). Our S&P materials EPS model inputs include the U.S. currency, metals commodity prices and a measure of borrowing costs. The model has been steadily decelerating recently, and moving in the opposite direction compared with sell-side analysts' optimistic estimates (bottom panel, Chart 5). Consequently, there is scope for downward revisions. Materials stocks are reflationary beneficiaries and also high fixed cost high operating leverage deep cyclicals that benefit most during the later stages of the business cycle when a virtuous capex/EPS upcycle takes root. A number of both developed and developing central banks have recently embarked on tightening monetary policy following in the Fed's footsteps. Global liquidity is on the verge of getting mopped up as even the ECB and the BoJ have started to hint that they would remove some of their ultra-accommodative and unconventional policy measures. These opposing forces keep us at bay and we continue to recommend a benchmark allocation in the S&P materials index. Chart 13Materials (Neutral) Real Estate (Neutral) Commercial real estate loan demand, a labor market measure and the EUR/USD comprise our S&P real estate profit growth model (second panel, Chart 14). The 10-year Treasury yield and real estate relative performance have been nearly perfectly inversely correlated since the GFC as REITs sport a hefty dividend yield and thus are considered a fixed income proxy. BCA's higher interest rate 2018 theme suggests that more downside looms for this rate-sensitive sector. Similarly, a firming EUR/USD reflecting the nearly 100% domestic exposure of the sector weighs on real estate relative performance. Our EPS model has recently sunk into the contraction zone and is in sync with sell-side analysts' negative profit growth figures for calendar 2018 (second panel, Chart 5). While all this signals that an underweight stance is appropriate, we would rather stay on the sidelines for three reasons: First, sector pricing power (mostly rents) has not eroded yet, despite the surge in multi-family housing construction. Second, most of the bad news is likely already discounted in sinking valuations and extremely oversold technicals. Finally, we would rather concentrate our interest rate related underweight in the pure play fixed income proxy, the utilities sector (please see page 15). Stick with a benchmark allocation in the S&P real estate index. Chart 14Real Estate (Neutral) Health Care (Underweight) Our S&P health care EPS growth model consists of health care pricing power, labor costs and a measure of health care outlays. Health care demand is fairly inelastic, signaling that health care spending prospects remain upbeat, especially given the aging population. However, the industry's up-to-recently structurally robust pricing power backdrop is under intense scrutiny. Medical commodity cost inflation is melting and drug pricing power has nearly halved since early 2016. Democrats and Republicans alike, despise the pharmaceutical/biotech industry's pricing tactics and drug price containment is on nearly every legislator's agenda. Add on the generic drug inroads, and Big Pharma/biotech resilient profits appear vulnerable, weighing heavily on the sector's relative performance. From a secular perspective, there is scope for health care sector profit gains. Developing countries are only just starting to institute social "safety nets" that the developed world already has in place. Our profit model is decelerating (second panel, Chart 15) and forecasting single digit EPS growth, in line with the Street's 12-month forward profit estimates (fourth panel, Chart 4). The S&P health care sector is a core underweight portfolio holding and we reiterate the high-conviction underweight status in the heavy weight S&P pharma sub index. Chart 15Health Care (Underweight) Utilities (Underweight) Utilities pricing power, the yield curve and analysts' net earnings revisions are the key inputs in our S&P utilities EPS growth model (second panel, Chart 16). While natgas prices, the industry's marginal price setter, have been stuck in a trading range between $2.6 and $3.4/mmbtu over the past 18 months, they are currently contracting and weighing heavily on industry pricing power. The U.S. economy is firing on all cylinders (bottom panel, Chart 16) and a selloff in the 10-year Treasury market near 3% is BCA's base-case scenario for 2018. Under such a backdrop, fixed income proxied defensive equities lose their luster, and thus utilities stocks will likely remain under intense downward pressure, Our S&P utilities EPS growth model is expanding at a mid-single digit growth rate, broadly in line with sell-side analysts' forecasts (fifth panel, Chart 4) and roughly 700bps below the broad market. The S&P utilities sector is a high-conviction underweight. Chart 16Utilities (Underweight) Technology (Underweight - Upgrade Alert) Our three-factor global technology EPS growth model includes capex intentions, the trade-weighted U.S. dollar and sell-side analysts' net earnings revision ratio. While the tech sector is still largely considered a deep cyclical, we view it as more defensive. The majority of large capitalization tech companies are mature, cash rich, cash flow generating, dividend paying and high margin. Tech firms thrive in a deflationary backdrop as business models have been built to withstand the inherently disinflationary "creative destruction" process. BCA's interest rate view calls for an inflationary driven sell off in bonds for 2018, suggesting that investors avoid high-flying tech stocks. Weakness in basic resources explains most of the delta in cyclical capital outlays. Encouragingly, technology's share of the U.S. capex pie is making inroads rising to roughly 10% (bottom panel, Chart 17). Tech investment has been so abysmal for so long that it is hard to get any worse. In fact, it has started to improve both on an absolute and relative basis, as pent-up tech demand is being unleashed. Our synchronized global capex upcycle theme is gaining traction and the tech sector will continue to make gains at the expense of resource-related spending. Our global tech EPS model is forecasting modest double-digit growth in the coming quarters (second panel, Chart 17), largely aligned with sell-side analysts' profit growth expectations (fifth panel, Chart 5). On balance, we are putting the S&P tech sector on upgrade alert reflecting the capex tailwind offsetting the rising interest rate backdrop, and reiterate our capex-related high-conviction overweight in the S&P software sub-index. Chart 17Technology (Underweight-Upgrade Alert) 1 Please see BCA U.S. Equity Strategy Weekly Report, "SPX 3,000?," dated July 10, 2017, available at uses.bcaresearch.com. 2 Please see BCA U.S. Equity Strategy Weekly Report, "EPS And "Nothing Else Matters"," dated December 18, 2017, available at uses.bcaresearch.com. 3 Please see BCA U.S. Equity Strategy Weekly Report, "Dissecting Profit Composition," dated July 24, 2017, available at uses.bcaresearch.com. 4 Please see BCA U.S. Equity Strategy Weekly Report, "Invincible," dated November 6, 2017, available at uses.bcaresearch.com. 5 Please see BCA U.S. Equity Strategy Weekly Report, "Dollar The Great Reflator," dated September 18, 2017, available at uses.bcaresearch.com. 6 Please see BCA U.S. Equity Strategy Weekly Report, "Can Easy Fiscal Offset Tighter Monetary Policy?," dated October 9, 2017, available at uses.bcaresearch.com.

Highlights An increase in the "synthetic" supply of bitcoins via financial derivatives, along with the launch of bitcoin-like alternatives by large established tech companies, will cause the cryptocurrency market to collapse under its own weight. Other areas that could see supply-induced pressures over the coming years include oil, high-yield debt, global real estate, and low-volatility trades. In contrast, the U.S. stock market has seen an erosion in the supply of shares due to buybacks and voluntary delistings. Investors should consider going long U.S. equities relative to high-yield credit, while positioning for higher volatility. Such an outcome would be similar to what happened in the late 1990s, a period when the VIX and credit spreads were trending higher, while stocks continued to hit new highs. A breakdown in NAFTA talks remains the key risk for the Canadian dollar and Mexican peso. Feature Bubbles Burst By Too Much Supply The "cure" for higher prices is higher prices. The dotcom and housing bubbles did not die fully of their own accord. Their demise was expedited by a wave of new supply hitting the market. In the case of the dotcom bubble, a flood of shares from initial and secondary public offerings inundated investors in 2000 (Chart 1). This put significant downward pressure on the prices of internet stocks. The housing boom was similarly subverted by a slew of new construction - residential investment rose to a 55-year high of 6.6% of GDP in 2006 (Chart 2). Chart 1Burst By Too Much Supply: Example 1 Chart 2Burst By Too Much Supply: Example 2 Is bitcoin about to experience a similar fate? On the surface, the answer may seem to be "no." As more bitcoins are "mined," the computational cost of additional production rises exponentially. In theory, this should limit the number of bitcoins that can ever circulate to 21 million, about 80% of which have already been created (Chart 3). Yet if one looks beneath the surface, bitcoin may also be vulnerable to a variety of "supply-side" factors. Chart 3Bitcoin: Most Of It Has Been Mined First, the expansion of financial derivatives tied to the value of bitcoin threatens to create a "synthetic" supply of the cryptocurrency. When someone writes a call option on a stock, the seller of the option is effectively taking a bearish bet while the buyer is taking a bullish bet. The very act of writing the option creates an additional long position, which is exactly offset by an additional short position. Moreover, to the extent that a decision to sell a particular call option will depress the price of similar call options, it will also depress the underlying price of the stock. This is simply because one can have long exposure to a stock either by owning it outright or owning a call option on it. Anything that hurts the price of the latter will also hurt the price of the former. As bitcoin futures begin to trade, investors who are bearish on bitcoin will be able to create short positions that cause the effective number of bitcoins in circulation to rise. This will happen even if the official number of bitcoins outstanding remains the same. Imitation Is The Sincerest Form Of Flattery An increase in synthetic forms of bitcoin supply is one worry for bitcoin investors. Another is the prospect of increased competition from bitcoin-like alternatives. There are now hundreds of cryptocurrencies, most of which use a slight variant of the same blockchain technology that underpins bitcoin. Chart 4Governments Will Want Their Cut So far, the proliferation of new currencies has been largely driven by technologically savvy entrepreneurs working out of their bedrooms or garages. But now companies are getting in on the act. The stock price of Kodak, which apparently is still in business, tripled earlier this week when it announced the launch of its own cryptocurrency. That's just a small taste of what's to come. What exactly is stopping giants such as Facebook, Amazon, Netflix, and Google from issuing their own cryptocurrencies? After all, they already have secure, global networks. Amazon could start giving out a few coins with every sale, and allow shoppers to purchase goods from the online retailer using its new currency. It's simple.1 The only plausible restriction is a legal one: The threat that governments will quash upstart cryptocurrencies for fear that will drive down demand for their own fiat monies. As we noted several weeks ago, the U.S. government derives $100 billion per year in seigniorage revenue from its ability to print currency and use that money to buy goods and services (Chart 4).2 As large companies get into the cryptocurrency arena, governments are likely to respond harshly - sooner rather than later. This week's news that the South Korean government will consider banning the trading of cryptocurrencies on exchanges is a sign of what's to come. Who Else? What other areas are vulnerable to an eventual tsunami of new supply? Four come to mind: Oil: BCA's bullish oil call has paid off in spades. Brent has climbed from $44 last June to $69 currently. Further gains may not be as easily attainable, however. Our energy strategists estimate that the breakeven cost of oil for U.S. shale producers is in the low-$50 range.3 We are now well above this number, which means that shale supply will accelerate. This does not mean that prices cannot go up further in the near term, but it does limit the long-term potential for crude. Real estate: Ultra-low interest rates across much of the world have fueled sharp rallies in home prices. Inflation-adjusted home prices in Canada, Australia, New Zealand, and parts of Europe are well above their pre-Great Recession levels (Chart 5). U.S. real residential home prices are still below their 2006 peak, but commercial real estate (CRE) prices have galloped to new highs (Chart 6). Rent growth within the U.S. CRE sector is starting to slow, suggesting that supply is slowly catching up with demand (Chart 7). Chart 5Where Low Rates Have ##br##Fueled House Prices Chart 6Commercial Real Estate Prices Have ##br##Surpassed Pre-Recession Levels Chart 7Rent Growth Is Cooling Corporate debt: Low rates have also encouraged companies to feast on credit. The ratio of corporate debt-to-GDP in the U.S. and many other countries is close to record-high levels (Chart 8A and Chart 8B). Credit spreads remain extremely tight, but that may change as more corporate bonds reach the market. Chart 8ACorporate Debt-To-GDP ##br##Is Close To Record Highs Chart 8BCorporate Debt-To-GDP ##br##Is Close To Record Highs Low-volatility trades: A recent Bloomberg headline screamed "Short-Volatility Funds Are Being Flooded With Cash."4 The number of volatility contracts traded on the Cboe has increased more than tenfold since 2012. Net short speculative positions now stand at record-high levels (Chart 9). Traders have been able to reap huge gains over the past few years by betting that volatility will decline. The problem is that if volatility starts to rise, those same traders could start to unload their positions, leading to even higher volatility. In contrast to the aforementioned areas, the stock market has seen an erosion in the supply of shares due to buybacks and voluntary delistings. The S&P divisor is down by over 8% since 2005. The number of U.S. publicly-listed companies has nearly halved since the late 1990s (Chart 10). This trend is unlikely to reverse any time soon, given the elevated level of profit margins and the temptation that many companies will have to use corporate tax cuts to step up the pace of share repurchases. Chart 9Low Volatility Is In High Demand Chart 10Erosion Of Supply In The Stock Market Bet On Higher Equity Prices, But Also Higher Volatility And Higher Credit Spreads The discussion above suggests that the relationship between equity prices and both volatility and credit spreads may shift over the coming months. This would not be the first time. Chart 11 shows that the VIX and credit spreads began to trend higher in the late 1990s, even as the S&P 500 continued to hit new record highs. We may be entering a similar phase now. Continued above-trend growth in the U.S. and rising inflation will push up Treasury yields. We declared "The End Of The 35-Year Bond Bull Market" on July 5, 2016 - the exact same day that the 10-year Treasury yield hit a record closing low of 1.37%.5 Higher interest rates will punish financially-strapped borrowers, leading to wider credit spreads. Equity volatility is also likely to rise as corporate health deteriorates and the timing of the next downturn draws closer. Our baseline expectation is that the U.S. and the rest of the world will fall into a recession in late 2019. Financial markets will sniff out a recession before it happens. However, if history is any guide, this will only happen about six months before the start of the recession (Table 1). This suggests that global equities can continue to rally for the next 12 months. With this in mind, we are opening a new trade going long the S&P 500 versus high-yield credit. Chart 11Volatility Can Increase And Spreads ##br##Can Widen As Stock Prices Rise Table 1Too Soon To Get Out Four Currency Quick Hits Four items buffeted currency and fixed-income markets this week. The first was a news story suggesting that China will slow or stop its purchases of U.S. Treasury debt. China's State Administration of Foreign Exchange (SAFE) decried the report as "fake news." Lost in the commotion was the fact that China's holdings of Treasurys have been largely flat since 2011 (Chart 12). China still has a highly managed currency. Now that capital is no longer pouring out of the country, the PBoC will start rebuilding its foreign reserves. Given that the U.S. Treasury market remains the world's largest and most liquid, it is hard to see how China can avoid having to park much of its excess foreign capital in the United States. The second item this week was the Bank of Japan's announcement that it will reduce its target for how many government bonds it buys. This just formalizes something that has already been happening for over a year. The BoJ's purchases of JGBs have plunged over the past twelve months, mainly because its ¥80 trillion target is more than double the ¥30-35 trillion annual net issuance of JGBs (Chart 13). Chart 12China's Holdings Of Treasurys: ##br##Largely Flat Since 2011 Chart 13BoJ Has Been Reducing ##br##Its Bond Purchases Ultimately, none of this should matter that much. The Bank of Japan can target prices (the yield on JGBs) or it can target quantities (the number of bonds it owns), but it cannot target both. The fact that the BoJ is already doing the former makes the latter irrelevant. And with long-term inflation expectations still nowhere near the BoJ's target, the former is unlikely to change. What does this mean for the yen? The Japanese currency is cheap and its current account surplus has swollen to 4% of GDP (Chart 14). Speculators are also very short the currency (Chart 15). This increases the likelihood of a near-term rally, as my colleague Mathieu Savary flagged this week.6 Nevertheless, if global bond yields continue to rise while Japanese yields stay put, it is hard to see the yen moving up and staying up a lot. On balance, we expect USD/JPY to strengthen somewhat this year. Chart 14Yen Is Already Cheap... Chart 15...And Unloved The third item was the revelation in the ECB's December meeting minutes that the central bank will be revisiting its communication stance in early 2018. The speculation is that the ECB will renormalize monetary policy more quickly than what the market is currently discounting. If that were to happen, EUR/USD would strengthen further. All this is possible, of course, but it would likely require that euro area growth surprise on the upside. That is far from a done deal. The euro area economic surprise index has begun to edge lower, and in relative terms, has plunged against the U.S. (Chart 16). Unlike in the U.S., the euro area credit impulse is now negative (Chart 17). Euro area financial conditions have also tightened significantly relative to the U.S. (Chart 18). Chart 16Euro Area Economic ##br##Surprises Edging Lower Chart 17Negative Credit Impulse In The Euro ##br##Area Will Weigh On Growth Chart 18Diverging Financial Conditions ##br##Favor U.S. Over The Euro Area Meanwhile, EUR/USD has appreciated more since 2016 than what one would expect based on changes in interest rate differentials (Chart 19). Speculative positioning towards the euro has also gone from being heavily short at the start of 2017 to heavily long today (Chart 20). Reasonably cheap valuations and a healthy current account surplus continue to work in the euro's favor, but our best bet is that EUR/USD will give up some of its gains over the coming months. Chart 19The Euro Has Strengthened More Than ##br##Justified By Interest Rate Differentials Chart 20Euro Positioning: From Deeply ##br##Short To Record Long Lastly, the Canadian dollar and Mexican peso came under pressure this week on news reports that the U.S. will be pulling out of NAFTA negotiations. Of the four items discussed in this section, this is the one that worries us most. The global supply chain has become highly integrated. Anything that sabotages it would be greatly disruptive. At some level, Trump realizes this, but he also knows that his base wants him to get tough on trade, and unless he does so, his chances of reelection will be even slimmer than they are now. Ultimately, we expect a new NAFTA deal to be reached, but the path from here to there will be a bumpy one. Housekeeping Notes Our long global industrials/short utilities trade is up 12.4% since we initiated it on September 29. We are raising the stop to 10% to protect gains. We are also letting our long 2-year USD/Saudi Riyal forward contract trade expire for a loss of 2.9%. Given the recent improvement in Saudi Arabia's finances, we are not reinstating the trade. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 My thanks to Igor Vasserman, President of SHIG Partners LLC, for his valuable insights on this topic. 2 Please see Global Investment Strategy Special Report, "Bitcoin's Macro Impact," dated September 15, 2017; and Global Investment Strategy Weekly Report, "Don't Fear A Flatter Yield Curve," dated December 22, 2017. 3 Please see Energy Sector Strategy Weekly Report, "Breakeven Analysis: Shale Companies Need ~$50 Oil To Be Self-Sufficient," dated March 15, 2017. 4 Dani Burger, "Short-Volatility Funds Are Being Flooded With Cash," Bloomberg, November 6, 2017. 5 Please see Global Investment Strategy Special Alert, "End Of The 35-year Bond Bull Market," dated July 5, 2016. 6 Please see Foreign Exchange Strategy, "Yen: QQE Is Dead! Long Live YCC!" dated January 12, 2018. Tactical Global Asset Allocation Recommendations Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights Environmental reforms in China continue to reduce steelmaking capacity. The shuttering of illegal induction facilities in China also is tightening markets. Although official Chinese steel output is higher, this likely reflects the fact that output from illegal induction mills went unreported - and thus uncounted - while production from legal mills is increasing to fill the gap left by closures. Steelmakers' profits are surging, which means demand for iron ore in China will remain stout at least through 1H18. Copper has been well bid since June 2017, following supply disruptions and strong demand growth driven by the global economic upturn. We expect it will get an additional lift in 1H18, as wiring and plumbing in construction projects now absorbing steel in China get underway. Later, global growth will make up for any slowdown in China. Our analysis indicates the global steel market will be tightening in 1H18, as it already is doing in China. Consistent with this, we are opening a tactical long position in Mar/18 steel rebar futures on the Shanghai Futures Exchange, which are quoted in RMB/ton. We are including a 10% stop loss on this recommendation. Energy: Overweight. Our once-out-of-consensus oil view is now the consensus, so we are taking profits on Brent and WTI $55 vs. $60/bbl call spreads on May- and July-delivery oil at tonight's close. These positions were up 109.2% and 123.5% at Tuesday's close. Any sell-offs will present an opportunity to re-establish length along these forward curves. Base Metals: Neutral. Copper will remain well bid this year as the global economic recovery rolls on. A large number of contract renegotiations at mines is an additional upside risk to copper prices this year. Precious Metals: Neutral. Given our expectation of four rate hikes by the Fed, it is difficult to get too bullish gold. However, any indication the central bank is tilting dovish - particularly if we fail to see higher inflation this year - will rally the metal. Ags/Softs: Underweight. Markets will tread water until Friday's USDA WASDE. We remain underweight, except for corn. Feature Chart of the WeekIron Ore And Steel Prices Diverged In 2017 China's environmental policy actions have reduced world steel-making capacity by 100 mm MT between 1H16 and 1H18. This is most visible in Chinese steel prices, which gained more than 30% in 2017, following an almost 80% increase in 2016. The total gain in steel prices since the start of Beijing's focus on steel-market reforms is a resounding 135%. Iron ore prices posted similar gains to steel in 2016 but diverged sharply in 2017, slumping more than 40% between mid-March and mid-June - ending almost 8% lower year-on-year (yoy) (Chart of the week). Soaring steel prices pushed profit margins at Chinese mills higher, which, of course, fed through to demand for iron ore, the critical steel-making ingredient in China, toward year-end: Iron ore prices were up 20% in the last two months of 2017. How Did We Get Here? A Recap Of China's Steel Sector Reforms As part of its reforms aimed at reducing air pollution by eliminating outdated, excess industrial capacity, Beijing pledged to eliminate 100-150 mm MT of steel capacity over the 2016-2020 period. To date it has shuttered an estimated 100 mm MT of capacity. In addition to these reforms Beijing pledged to shut down smaller induction furnaces in China, which melt scrap steel, and produce steel of shoddy quality. These induction furnaces are estimated to account for 80-120 mm MT worth of annual capacity, although their actual output is far less: They produced an estimated 30-50 mm MT in 2016, according to S&P Global Platts.1 This is less than 7% of China's total crude-steel output. Production cuts from induction mills are not evident in official data - China's crude steel production figures have continued to rise amid these cuts, as we discussed in previous research (Chart 2).2 Data from the International Iron and Steel Institute shows global steel output was at a record high for the first 11 months of 2017, increasing by more than 5% yoy. Likewise, crude steel output from China - which accounts for 50% of global output - peaked in August: Output over the same period was the highest on record, increasing by 5.28% compared to the same period in 2016. This production paradox can be put down to the fact that many Chinese induction furnaces are illegal, and, as a result their output is not accounted for in official production data. As legal steelmakers ramped up their output to offset declines from the closed down induction furnaces, official crude production figures climbed. In fact, further examination of Chinese steel data makes it clear that China's steel market is in fact tighter than what can be inferred from the crude production figures (Chart 3). The following observations point to a strained market: While China's crude steel production has been paving new record highs, China Stat Info data reveals a contradictory picture about steel products. Output of steel products in the March to November period of 2017 came in 3.46% lower yoy, marking the first yoy decline for that period since 1995! While crude steel produced by induction furnaces would not be included in official crude steel figures, the metal would eventually be used to manufacture steel products - wires, rods, rails and bars, and are represented in this data. Thus the decline in steel products indicates that lower crude supply has weighed down on the output of steel products. China's steel exports have been on a downtrend. In theory, this can be due to either an increase in domestic demand or a decrease in foreign demand. Given the healthy state of the global economy, and what we know about steel production in China, we are believers in the former theory. China's exports of steel products are down 30% yoy in the first 11 months of 2017. Aside from the 3.04% yoy decline in 2016, these mark the first annual declines in exports since 2009. In face of lower domestic supply, China has likely reduced its exports in order to satisfy demand from local steel users. China's scrap steel imports fell in 2H17. Unlike blast furnaces which use iron ore as the main input in steelmaking, the shuttered illegal steelmakers use scrap steel which they melt in an induction furnace. Coincident with the elimination of these furnaces, China's imports of scrap steel fell 14.35% yoy in 2H17. This is further evidence of reduced demand for the scrap steel from these furnaces. China steel inventories are falling. In fact steel product inventories in major industrial cities are at record lows (Chart 4). This is a symptom of a tight market with demand outpacing supply, contradicting China's crude steel production figures. Chart 2Record Chinese Production Of Crude Steel##BR##Amid Falling Steel Products Output Chart 3China Trade Data Evidence##BR##Of Tight Market Chart 4Steel Inventories##BR##In China Are Falling Furthermore, according to World Steel Association (WSA), capacity utilization in the 66 countries for which they collect data increased by 3.12 percentage points yoy for the July to November 2017 period to average 72.64%, up from the 69.52% average in the same period of 2016. These observations are evidence that despite the increase in official crude steel production figures, the actual output has in fact fallen and supply is tighter. Whether steel prices will remain buoyed by tight supply hinges on whether China is successful in permanently shuttering excess capacity and shoddy steel producers, or if induction furnace operators are able to circumvent these policies and bring illegal steel back to the market. China's Reforms To Dominate Steel Market, At Least This Winter Following the conclusion of the mid-December Central Economic Work Conference, Chinese authorities announced the "three tough battles" for the next three years, which they see as crucial for future economic prosperity. These battles are summarized as (1) preventing major risks, (2) targeted poverty alleviation, and (3) pollution control. The International Energy Agency (IEA) estimates that air pollution has led to ~1 million premature deaths while household air pollution caused an additional 1.2 million premature deaths in China annually.3 Because of this, improving China's air quality is a chief social and health target for China. Chart 5Lower Chinese Steel Production##BR##Will Impact Global Output This will mean that measures to reduce pollution and clear China's skies will be critically important to the steel sector. According to the Ministry of Environmental Protection, China has pledged a 15% yoy reduction in the concentration of airborne particles smaller than 2.5 microns in diameter - known as PM2.5 - in 28 smog-prone northern cities. The steel industry, which is mostly concentrated in the northern China region of Beijing-Tianjin-Hebei, is one of the top sources of air polluting emissions in that region. In fact, industrial emissions - most notably from the steel and cement sectors - are reportedly responsible for 40-50% of these small airborne particles. China's winter smog "battle plan" will target these polluting industries by mandating cuts on steel, cement, and aluminum production during the smog-prone mid-November to mid-March months, as well as restricting household coal use, diesel trucks and construction projects. Steel production cuts target a range between 30-50%, which, according to Platts estimates, will take 33 mm MT of steel production - equivalent to ~3.9% of China's projected 2017 crude steel output - offline during the winter. In fact, according to China's environment minister, Li Ganjie, "these special campaigns are not a one-off, instead it is an exploration of long-term mechanisms."4 Thus, these cuts may become a recurring event in China's steel sector. China's official crude steel figures are beginning to show the impact of these cuts with November crude production falling 8.6% month-on-month (mom) and growing by just 2.2% yoy - significantly slower than the 7.6% yoy average experienced since July. As a consequence, although crude production in the rest of the world grew in line with previous months, global steel output fell almost 6% month-on-month in November, while yoy production grew 3.7% – a significant deceleration from the average 6.6% yoy rate witnessed since the beginning of 2H17 (Chart 5). Risks to this outlook come from weak compliance with these cuts. There are recent reports of evasions by aluminum and steel producers in Shandong. Nonetheless, given China's focus on these reforms, we do not foresee widespread violations. Another risk comes from the demand side. As part of its environmental agenda, Beijing announced plans put off the construction of major public projects in the city - road and water projects - until springtime. The suspension is not intended to impact "major livelihood projects" such as railways, airports, and affordable housing. Construction is the largest end user for steel - according to WSA more than half of global steel is used for buildings and infrastructure - a slowdown in the construction sector would weigh on steel demand.5 If other major construction zones adopt a similar policy, the impact of lower steel supply will be offset by weak demand, muting the overall effect on the steel market. Bottom Line: We expect to see lower steel production and exports from China in the coming months. Given Xi Jinping's resolve to improve air quality, we expect compliance to environmental reforms among steelmakers to be strong this winter. This, along with lower output from induction furnaces in China, indicates the market could be tighter than is commonly supposed at least in 1H18. The likelihood the global economic recovery and expansion persists through 2018 suggests steel markets could remain well bid in 2H18, particularly if, as we expect, growth ex-China picks up the slack resulting from any slowdown in China. However, we will need to see what the actual reforms for the industry look like following the National People's Congress in March 2018.6 Steel Profit Margins Spur Iron Ore Demand Given steel's exceptional price gains over the past two years, and iron ore's lackluster performance in 2017, profit margins at China's steel producers reached multi-year highs (Chart 6). Ordinarily, this would normally encourage steel production, which would flood the market with supply and push prices down. However, China's environmental reforms will cap output from the country's most productive steelmaking region in coming months. Consequently, unless there are mass policy violations by steel producers this winter, we do not anticipate a swift price adjustment lower. Instead, steel producers are preparing to run on all cylinders when production restrictions are lifted in the spring. As such, they are filling iron ore inventories and taking advantage of weaker iron ore prices, before the iron ore market catches up with steel. China's iron ore imports reached an all-time record in September, while the latest data shows a 19% month-on-month (mom) jump in imports, corresponding with a 2.8% yoy increase (Chart 7). Chart 6Healthy Steel##BR##Profit Margins Chart 7Steel Producers Stocking Up On Iron Ore##BR##In Preparation For Spring This runs counter to what we expect during a period of muted steel production. Especially in an environment of healthy iron ore inventories, as China is in currently. Although Chinese inventories came down from mid-year peaks, they resumed their upward trend in 4Q17. This coincides with the steel winter capacity cuts, and is likely due to reduced demand for the ore from steel mills. There are two theories to explain this phenomenon: 1. Chinese steelmakers are taking advantage of lower iron ore prices and locking in higher profit margins, in anticipation of higher iron ore prices once steel production picks up again in the spring. 2. Amid the winter cuts, China's steelmakers are demanding high-grade iron ore, imported from Brazil and Australia. This will help them ensure that they are able to maximize their output without violating environmental policies. Environmental Consciousness Widens Iron Ore Spreads A consequence of the steel winter capacity cuts is stronger demand for higher grade raw materials to cut down on the most polluting phases of steel production. Higher-grade iron ore, which is defined by its purity or iron content, is more efficient for blast furnaces to use, allowing them to produce more steel from each tonne of iron ore they consume, maximizing output and profit. This is especially true in a tight steel market, with healthy profit margins: Steelmakers are able to afford the higher grades and are favoring productive efficiency. The discount for lower grade iron ore fines - 58% iron content - as well as the premium for higher grade 65% iron content have widened (Chart 8). This is because mills have found a way to legally circumvent the winter environmental restrictions, and still remain compliant. Furthermore, purer ores are less polluting, which helps serve China's environmental agenda. In addition, the premiums for iron ore pellets and iron ore lumps have also widened. Unlike lumps and pellets which can be fed directly into blast furnaces, fines require a sintering process which is highly polluting. Thus, China's environmental reforms have increased demand for higher grade, less polluting ores. An additional factor that could be driving up spreads is higher metallurgical coke prices (Chart 6). Higher grade iron ore contains less silica and thus requires less met coke to purify the ores. According to anecdotal evidence from China, Carajas fines from Brazil - which have the highest iron ore content and lowest silica content- are reportedly in high demand.7 Furthermore, China's imports show a decline in iron ore from India - which is of the lower grades. In the July to October period, imports fell 11.26% yoy with October imports falling almost 25% yoy and 30% mom. This is consistent with the theory that steel makers are shunning lower grade ores. On the other hand, imports from Brazil and Australia are expected to remain strong (Chart 9). The latest Australian Resources and Energy Quarterly forecasts Australian and Brazilian iron ore exports to grow 5.4% and 4.2% respectively in 2018, while Indian exports are projected to fall 57.5% yoy. Chart 8Wide Iron Ore##BR##Price Spreads Chart 9Environmental Concerns Will Support##BR##Demand For High Grade Iron Ore Bottom Line: In an effort to keep production high and profit from strong steel prices in face of the winter production cuts, steel producers are turning to higher-grade iron ore, pushing up the spread between high vs. low grade ores. The extent to which steel producers are able to successfully keep production going on the back of higher-grade ores will dampen the impact of the winter production cuts on the steel sector. Given that China's environmental focus is a long term plan, we expect these spreads to remain wide, rather than reverting back to their historic average. Steel Prices And Copper Markets Chart 10Steel Consumption Helps##BR##Predict Copper Prices The copper market had a roller coaster fourth quarter. Prices for the red metal were on a general uptrend since May, and first peaked in early September at $3.13/lb before bottoming at $2.91/lb by the second half of that month. Shortly thereafter, copper prices peaked at a new high of $3.22/lb by mid October - their highest in more than three years. Fears of a slowdown in China following messaging from the 19th Communist Party Congress caused the metal to lose almost 10% of its value, when it bottomed for the second time in early December. In fact, this coincided with a 4.65% decline in the price on December 5. While there is no clear justification for this fall, it can be put down to a mix of factors including a ~10 th MT increase in LME inventories, worries about a China slowdown, as well as a liquidation of positions ahead of the new year. Nonetheless, copper has since regained these losses to end the year at $3.28/lb. In our modelling of copper, we find that steel consumption is significant in forecasting future copper price behavior. More specifically, China's steel consumption has a significant positive relationship with copper prices 6 months into the future (Chart 10). This can be explained by the importance of the construction sector as an end user of both materials. However, each metal goes into the construction site at different time frames. While steel products are used in the construction of the structures, and thus are needed at the beginning of the project, copper is used in the electrical wiring and plumbing, and is thus needed later (6 months or so) in the project. This is in line with our findings that steel is most significant with a six-month lag - reflecting the average time period between which the structure is built and the plumbing and wiring are needed. Steel consumption in China is a useful leading indicator of copper markets when demand side fundamentals are dominating steel and copper markets. Government stimulus and a solid construction sector boosted China's steel demand in 2017. However, according to the WSA Short Range Outlook, demand for steel will moderate this year on the back of reflation in China, partially offset by strong global growth. WSA notes that the closure of induction furnaces skewed up steel demand growth figures to 12.4% yoy, and instead cite a more reasonable estimate along the lines of 3% yoy steel demand growth from China in 2017, bringing the global steel demand growth rate to 2.8%. While steel demand outside of China grew by an estimated 2.6% in 2017, they foresee it reaching 3% in 2018. In contrast, they expect flat demand from China in 2018, bringing world steel demand growth to 1.6% in 2018 (Table 1). Table 1Steel Demand (yoy Growth Rates) Moderating demand from China and the stability (or lack thereof) of the supply-side will dominate the copper market this year. On the demand side, China's steel market offers insight about the future direction of the red metal. Bottom Line: Given China's appetite for steel has remained healthy to date and is projected to maintain its 2017 level this year, we do not expect a demand-induced plunge in copper prices in the 6 month horizon. Roukaya Ibrahim, Associate Editor Commodity & Energy Strategy RoukayaI@bcaresearch.com 1 Please see "Will China's induction furnace steel whac-a-mole finally come to an end?" published by S&P Global Platts March 6, 2017. 2 Please see BCA Research's Commodity & Energy Strategy Weekly Report "Slow-Down In China's Reflation Will Temper Steel, Iron Ore In 2018," published September 7, 2017, available at ces.bcaresearch.com. 3 Please see IEA World Energy Outlook 2016 Special Report titled "Energy and Air Pollution," available at iea.org. 4 Please see "Provincial China officials used fake data to evade aluminium, steel capacity curbs - China Youth Daily," published on December 26, 2017, available at reuters.com. 5 Please see "Steel Markets" at worldsteel.org. 6 For additional discussion, please see "Shifting Gears in China: The Impact On Base Metals," in the November 9, 2017, issue of BCA Research's Commodity & Energy Strategy. It is available at ces.bcaresearch.com. 7 Please see "High-medium grade iron ore fines spread widens to all-time high of $23.55/dmt," published August 22, 2017, available at platts.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed in 2017

Special Report

Highlights The beta of Chinese stocks has been steadily increasing over the past few years, versus both emerging markets and global stocks. Rising relative currency volatility has likely durably increased the cyclicality of Chinese stock prices. The high-beta nature of Chinese investable stocks suggests that they should be favored when the EM and global stock benchmarks are rising. This supports our current overweight stance. A portfolio strategy that favors equity sectors with high alpha significance has outperformed the broad investable market by a non-trivial amount over time, without adding to portfolio risk. Barring a few exceptions, the model's current allocation is generally consistent with our theme of a benign slowdown in Chinese economic growth. Feature Chart 1Beta Matters, But So Does Alpha While concepts such as alpha, beta, and correlation are frequently applied by investment managers at the security or sector level, they are less commonly employed from a top-down regional equity perspective and are rarely examined as a time series. In addition, the concept of alpha persistence (i.e. alpha that is persistently positive or negative) is also frequently ignored by investors, despite it having significant implications for portfolio returns. This is vividly illustrated by the relative performance of developed commodity markets during the last economic expansion: these countries resoundingly outperformed a rising global benchmark from 2000 to 2007, despite having a market beta that averaged one over the period (Chart 1). This seeming inconsistency is explained by persistent volatility-adjusted outperformance throughout the period (panel 3), underscoring the importance of tracking this measure from a top-down perspective. In this report we examine the recent evolution of MSCI China's alpha and beta versus both the emerging market (EM) and global benchmarks. We conclude that China is no longer a low-beta market (supporting an overweight stance), and also present a simple alpha-based sector model for Chinese investable stocks that has generated impressive outperformance over time without adding to portfolio risk. The Evolution Of China's Alpha & Beta Chart 2 presents the evolution of alpha and beta for Chinese investable stocks since 2010, versus the emerging market and global index. Given the significant outperformance of the technology sector over the past year, we also present this analysis in ex-tech terms. The values shown in Chart 2 are calculated using a standard single-factor model approach to estimating alpha and beta, namely a regression of weekly stock price returns in US$ terms in excess of the return from U.S. short-term Treasury bills on excess returns of the benchmark index.1 The chart yields the following observations: The beta of Chinese stocks has been steadily increasing over the past few years, versus both emerging markets and global stocks, regardless of whether the tech sector is removed from the picture. Chinese stocks had a beta of 1.4 versus their global peers in 2017, placing it in the 80th percentile of all country equity market betas for the year. Chinese stocks earned a modestly negative alpha vs global stocks in 2016, which was even larger when compared to the EM benchmark. This likely occurred because of lower exposure to resource-oriented sectors, given the significant rebound in commodity prices in 2016. Chinese stocks experienced a surge in alpha in 2017, even excluding technology stocks. In 2017, in all cases (vs EM and global, including or excluding tech) Chinese equities moved into the top right alpha/beta quadrant, which is the quadrant that offers the highest return to investors when the benchmark is rising. This is a remarkable development given that there were indications of a peak in Chinese economic momentum in the first half of the year, and suggests that investors do not view the ongoing slowdown as being problematic for investable equity performance. Chart 2 raises the obvious question of why China has become a higher beta market. We have two theories, but only the second one appears to fit the data. The first theory is that the establishment of the stock connect in late-2014 caused a volatility spillover from China's domestic stock market into the investable market. But while it is true that A-shares were considerably riskier than investable stocks in late-2015 / early-2016, Chart 3 makes it clear that A-shares have not historically been much more volatile than investable stocks. In addition, Chart 2 underscores that the rise in China's market beta since 2014 has been persistent, whereas A-shares in 2017 recorded their lowest share price volatility in over 15 years. So to us, this does not appear to be the most probable explanation. Chart 2China Has Become A High-Beta Market The second theory, which seems much more likely, is that the rising currency volatility has increased the cyclicality of Chinese stock prices. China's decision to devalue the RMB in August 2015 clearly led to a period of significantly increased capital controls, but Chart 4 highlights that the CNY/USD exchange rate has steadily become more volatile. This is especially true when compared with a basket of emerging market currencies, with CNY/USD actually being more volatile than the basket over the past year. Chart 3The Stock Connect Does Not Explain##br## The Rise In China's Beta Chart 4Rising Relative Currency Volatility ##br##= Higher Beta While it is certainly true that Chinese policymakers have stepped up their management of the currency by tightening capital controls over the past year, the PBOC's decision to pursue its "partial" version of the impossible trinity still implies, in our view, that RMB volatility will now be structurally higher than what prevailed on average prior to August 2015.2 This suggests that China's equity market beta will be durably higher than before, absent a presently negative correlation between CNY/USD and EM or global stock prices. Bottom Line: The beta of Chinese stocks has been steadily increasing over the past few years, versus both emerging markets and global stocks. Rising relative currency volatility has likely durably increased the cyclicality of Chinese stock prices. Investment Implications Of China's Recent Relative Performance There are two clear investment strategy implications from Chinese equities becoming a high-beta asset. The first is that Chinese investable stocks are now a pro-risk asset to be favored when the EM and global stock benchmarks are rising. Chart 5 shows that both are currently well above their 200-day moving averages, which supports our overweight stance towards China. The second is that when comparing the performance of China's overall investable index versus that excluding technology, it is clear that a non-trivial amount of the alpha earned by China's overall index in 2017 came from the tech sector. This suggests that a reversal of the high-flying performance of Chinese technology stocks is a material risk to our overweight stance towards Chinese equities. For now, this high-alpha outperformance appears to be fundamentally-based: Chart 6 highlights that forward earnings for Chinese tech shares have risen enormously relative to the investable benchmark over the past three years, a trend that we have noted appears to be driven by Chinese consumer demand (and thus unlikely to decline over the coming year).3 In addition, the relatively modest but positive alpha earned by Chinese ex-tech stocks in 2017 was likely driven by extremely cheap valuation, and these multiples remain quite low relative to other countries. We highlighted in our December 7 Weekly Report that the relative re-rating of Chinese investable ex-tech stocks was a key theme for 2018,4 suggesting that there is room for further re-rating/alpha if China's economic slowdown remains benign (as we expect). Chart 5Investors Should Overweight ##br##Chinese Stocks In This Environment Chart 6Tech's Recent Alpha Appears ##br##Fundamentally-Based Bottom Line: The now high-beta nature of Chinese investable stocks suggests that they are a pro-risk asset to be favored when the EM and global stock benchmarks are rising. This supports our current overweight stance. Alpha, Applied: A Simple Sector Model For Chinese Investable Stocks We noted earlier that the concept of alpha has had significant implications for regional equity portfolio returns in the past. In order to test the predictive power of alpha within the context of a Chinese equity portfolio, we evaluate the returns of an investment strategy that allocates to China's investable equity sectors based on the significance of alpha. Table 1 presents statistics summarizing the performance of this sector alpha portfolio relative to the overall investable market, Table 2 shows the portfolio's current sector allocation, and Chart 7 illustrates the cyclical behavior of the portfolio's relative performance trend since 2004. Several important conclusions emerge: Table 1An Alpha-Based Sector Model Has Historically Outperformed ##br##China's Investable Stock Market Table 2Sector Alpha Portfolio Weights Are Generally Consistent With ##br##A Benign Growth Slowdown The model has outperformed the broad investable market by an impressive 235 bps per year without appearing to take on any additional risk. Measured either as volatility or drawdown, the riskiness of the portfolio appears to be the same as that of the overall investable market. The outperformance of the model occurs in spurts, but sustained periods of underperformance are not common. The 2007-2009 period served as an exception to this rule, but even in this case the cumulative underperformance of the model vs the investable index was not large (roughly 6%). Chart 7Impressive Outperformance Over Time The model is currently underweight financials (significantly), energy, industrials, telecoms, and utilities. Overweights are concentrated in the tech sector, real estate, health care, and consumer stocks. For now, these weights are generally consistent with our benign slowdown scenario, although there are some potential exceptions to monitor (such as the overweight stance towards real estate and materials). Bottom Line: A portfolio strategy that favors equity sectors with high alpha significance has outperformed the broad investable market by a non-trivial amount over time, without adding to portfolio risk. Barring a few exceptions, the model's current allocation is generally consistent with our theme of a benign slowdown in Chinese economic growth. Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 2 Please see China Investment Strategy Weekly Report, "How Will China Manage The Impossible Trinity", dated December 8, 2016, available at cis.bcaresearch.com. 3 Please see China Investment Strategy Special Report, "The Data Lab: Testing The Predictability of China's Business Cycle", dated November 30, 2017, available at cis.bcaresearch.com. 4 Please see China Investment Strategy Weekly Report, "Three Themes For China In The Coming Year", dated December 7, 2017, available at cis.bcaresearch.com. Cyclical Investment Stance Equity Sector Recommendations

Highlights Chart 1Bond Bear On Pause? The start of a new year often brings optimism and nowhere is this more evident than in economic projections. In three of the past four years (2017 being the exception) Bloomberg consensus GDP growth expectations ended the year lower than where they began. A related pattern played itself out in the Treasury market. At the turn of each of the past four years the average yield on the Bloomberg Barclays Treasury Index increased in December only to fall back in January. In two of those instances the January decline exceeded the December increase. Should we expect a similar January bond rally this year? Our favorite short-term indicators are not sending a strong signal (Chart 1). Net speculative futures positions weakly suggest that the 10-year yield will be lower in three months, but our auto regressive model suggests the Economic Surprise Index will still be in positive territory at the end of the month. In a recent report we showed that yields tend to rise in months where the Surprise Index is above zero.1 Perhaps most importantly, our 2-factor Treasury model shows that yields are significantly lower than is suggested by global economic fundamentals. Maintain below-benchmark duration. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 49 basis points in December and by 335 bps in 2017. At 94 bps, the average index spread is 28 bps tighter than at the beginning of 2017 and investment grade corporate spreads are extremely expensive compared to history (Chart 2). After adjusting for changes in the average duration of the index over time, we calculate that A-rated corporate spreads have only been tighter 5% of the time since 1989 (panel 2), and Baa-rated spreads have only been tighter 7% of the time (panel 3). Essentially, at this stage of the credit cycle we should expect excess returns no greater than carry. As for the credit cycle itself, we noted in our last report that with corporate balance sheets deteriorating, low inflation and still-accommodative monetary policy are the sole supports for corporate spreads.2 We expect spreads will start to widen later this year once inflation rises and policy becomes more restrictive. With excess returns likely to be lower in 2018 than in 2017, we should also expect a lower marginal return from increasing the riskiness within credit portfolios.3 For investors looking to scale back on credit risk, our model shows that Financials and Technology are the most attractive low-risk sectors. Energy, Basic Industry and Communications are all attractive high-risk sectors (Table 3). Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Table 3BCorporate Sector Risk Vs. Reward* High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 23 basis points in December and by 602 bps in 2017. The average index option-adjusted spread tightened 1 bp on the month and 66 bps in 2017. Though spreads appear somewhat more attractive than for investment grade corporates, there is still not much room for spread compression in high-yield. In fact, we calculate that if the high-yield index spread tightens another 117 bps, junk bonds will be the most expensive they have been since 1995. In an optimistic scenario where the index spread tightens 100 bps, bringing it close to all-time expensive levels, then we would expect junk excess returns to be in the range of 600 bps (annualized). Given trends in corporate leverage, another 100 bps of spread tightening should be viewed as unlikely. More realistically, we expect excess returns in the range of 200 bps to 500 bps (annualized) between now and the end of the credit cycle (Chart 3). Given our forecast for default losses, flat spreads translate to a 12-month excess return of 213 bps. An additional warning sign for junk spreads is that the slope of the 2/10 Treasury curve is hovering around 50 bps. We showed in a recent report that when the 2/10 slope is between 0 bps and 50 bps, junk bonds underperform Treasuries in 48% of months, and average monthly excess returns (though still positive) are much lower than when the curve is steeper.4 MBS: Neutral Chart 4MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 16 basis points in December and by 51 bps in 2017. The conventional 30-year zero-volatility MBS spread narrowed 2 bps in December, the combination of a flat option-adjusted spread (OAS) and a 2 bps decline in the compensation for prepayment risk (option cost). The Z-spread widened 2 bps in 2017, as an 8 bps OAS widening was offset by a decline of 6 bps in the compensation for prepayment risk. The substantial OAS widening in early 2017 was almost certainly caused by investors pricing-in the eventual run-off of the securities on the Fed's balance sheet. Now that run-off has begun we see no obvious catalyst for further OAS widening in the months ahead. Turning to the compensation for prepayment risk, with Treasury yields biased higher as the Fed continues to lift rates, we see little risk of a material increase in refinancing activity. This will ensure that overall MBS spreads stay capped near historically low levels (Chart 4). All in all, with MBS OAS looking more attractive relative to Aaa-rated credit than at any time since 2015 (panel 3), we think this is an opportune time for investors looking to de-risk their portfolios to shift some of their spread product allocation away from corporate bonds and into MBS. We already upgraded our recommended allocation to MBS from underweight to neutral in October, and will likely further increase exposure as we advance toward the end of the credit cycle. Government-Related: Underweight Chart 5Government-Related Market Overview The Government-Related index underperformed the duration-equivalent Treasury index by 5 basis points in December, but outperformed by 216 bps in 2017. Sovereign bonds underperformed the Treasury benchmark by 36 bps in December, Foreign Agencies and Domestic Agencies underperformed by 8 bps and 1 bp, respectively. Local Authorities outperformed the benchmark by 17 bps, and Supranationals underperformed by 1 bp. Sovereign bonds were the best performers within the Government-Related index in 2017, delivering excess returns of 538 bps relative to duration-matched U.S. Treasuries. This outperformance was concentrated early in the year and was driven by the sharp depreciation of the U.S. dollar (Chart 5). With the market still priced for a relatively modest 63 bps of Fed rate hikes during the next 12 months, further sharp dollar depreciation appears unlikely. We recommend an underweight allocation to Sovereign debt. We remain overweight Local Authority and Foreign Agency bonds, sectors that delivered excess returns of 420 bps and 248 bps, respectively in 2017. Despite the outperformance, both of these sectors still offer attractive spreads after adjusting for credit rating and duration. We remain underweight Domestic Agency and Supranational bonds. Though both sectors offer low risk and high credit quality, they also only offer 15 bps and 17 bps of option-adjusted spread, respectively. We much prefer Agency-backed MBS and CMBS which are also relatively low risk and offer option-adjusted spreads of 28 bps and 42 bps, respectively. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 99 bps in December and by 332 bps in 2017 (before adjusting for the tax advantage). The average Aaa Municipal / Treasury (M/T) yield ratio fell 5% in December, and is 12% below where it began 2017 (Chart 6). The recent decline follows a sharp increase that was driven by fluctuating supply trends related to the passage of U.S. tax legislation. The final tax bill ends the practice of advance refunding municipal bonds. As a result, December set a new high of $55.6 billion for municipal issuance as issuers rushed to get their advance refunding deals to market before the bill was passed (panel 3). Now that the bill has passed, visible supply has evaporated and the average M/T yield ratio has fallen back to one standard deviation below its post-crisis mean. The absence of advance refunding will bias municipal bond issuance lower in 2018, thus removing one potential risk for yield ratios. The M/T yield ratio for short maturity debt has risen considerably relative to the yield ratio for long maturity debt in recent months (panel 2), and the risk/reward trade-off now appears more balanced. We close our recommendation to favor long maturities versus short maturities on the Aaa Muni curve. The third quarter update of our Muni Health Monitor showed a slight improvement (panel 5), but still no clear reversal of trend. Although health remains supportive for now - and consistent with municipal upgrades outpacing downgrades - with yield ratios close to their lows we maintain an underweight allocation to Municipal bonds.  Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview The Treasury curve bear-flattened in December. The 2/10 Treasury slope flattened 13 bps on the month, and the 5/30 Treasury slope flattened 15 bps. The evolution of the Treasury curve in 2018 will come down to a trade-off between how quickly inflation rises versus how quickly the Fed lifts rates. For example, in a recent report we showed that the 10-year Treasury yield will likely settle into a range between 2.80% and 3.25% by the time that core PCE inflation reaches the Fed's 2% target.5 That same report shows that if that adjustment occurs relatively quickly, and the Fed has only lifted rates once or twice between now and then, then the 2/10 Treasury slope is much more likely to steepen than to flatten. Conversely, if the Fed lifts rates three or four more times between now and the time that inflation returns to target, then the curve is more likely to flatten. For our part, we think it is wise to maintain a position long the 5-year bullet and short a duration-neutral 2/10 barbell. Such a position profits from a steeper curve, and our model shows that the butterfly spread is currently priced for significant curve flattening (Chart 7). According to our model, the 2/5/10 butterfly spread is discounting 27 bps of 2/10 flattening during the next six months.6 In other words, if the 2/10 slope steepens or flattens by less than 27 bps, then our recommended position will profit. TIPS: Overweight Chart 8TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 41 basis points in December, but underperformed by 43 bps in 2017. The 10-year TIPS breakeven inflation rate went on a wild ride last year. It started 2017 at 1.95% and, driven by strong inflation prints and continued post-election euphoria, reached as high as 2.09% in January. The breakeven dropped to a low of 1.66% in June, as inflation started to disappoint in the second quarter, but has rebounded during the past couple of months and just recently broke back above 2%. The 10-year TIPS breakeven rate is currently 2.02%, above where it began 2017. According to our TIPS Financial Model, the recent widening in breakevens is in line with the message from other related financial market instruments (Chart 8). Specifically, oil prices, the trade-weighted dollar and the stock-to-bond total return ratio. Further, measures of pipeline inflation pressure continue to signal an increase in inflationary pressures (panels 3 and 4), and the trimmed mean PCE shows that the realized inflation data are forming a tentative bottom (bottom panel). The annualized 6-month rate of change in the trimmed mean PCE ticked up to 1.68% in November, higher than the 12-month rate of change (1.67%). The 1-month rate of change is higher still at 2.19%, annualized. We continue to see signs that inflation will start to rebound in the coming months, and this will cause long-maturity TIPS breakeven inflation rates to reach a range between 2.4% and 2.5% by the time that inflation returns to the Fed's target. Remain overweight TIPS versus nominal Treasury securities. ABS: Neutral Chart 9ABS Market Overview Asset-Backed Securities performed in line with the duration-equivalent Treasury index in December and outperformed by 92 basis points in 2017. In 2017, Aaa-rated ABS outperformed the Treasury benchmark by 79 bps and non-Aaa ABS outperformed by 217 bps. The index option-adjusted spread for Aaa-rated ABS widened 1 bp in December, but tightened 21 bps in 2017. It now sits at 31 bps, only 4 bps above its all-time low (Chart 9). At 31 bps, Aaa-rated ABS now offer only a 3 bps spread advantage over Agency-backed MBS, and offer 11 bps less spread than Agency-backed CMBS. With consumer lending standards tightening and delinquency rates rising, we view no more than a neutral allocation to ABS as appropriate. On lending standards, the Fed's October Senior Loan Officer's Survey showed a continued tightening in lending standards on both credit cards and auto loans (panel 4), and also that demand for credit card and auto loans was essentially unchanged from the prior quarter. It also included a set of special questions regarding the reasons for changes in the supply and demand for consumer credit. Banks cited a less favorable or more uncertain economic outlook, a deterioration in existing loan quality and a general reduced risk tolerance as reasons for tightening the supply of credit. The hard data confirm that banks are seeing a deterioration in the quality of their consumer loan books (bottom panel). Although delinquencies remain depressed compared to history, with ABS spreads near all-time tights, rising delinquencies and tightening lending standards make for a poor risk/reward trade-off in the sector. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 20 basis points in December and by 201 bps in 2017. The index option-adjusted spread for non-agency Aaa-rated CMBS tightened 2 bps in December and 13 bps in 2017. At its current level of 64 bps, the index spread is about one standard deviation below its pre-crisis mean, and only 13 bps above its all-time low reached in 2004 (Chart 10). With spreads at such low levels in an environment of tightening commercial real estate (CRE) lending standards and falling CRE loan demand, we continue to view the risk/reward trade-off in non-Agency CMBS as unfavorable. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 21 basis points in December and by 133 bps in 2017. The index option-adjusted spread for Agency CMBS tightened 3 bps in December and 13 bps in 2017. At its current level of 42 bps, the sector offers greater option-adjusted compensation than a position in Agency-backed MBS (28 bps) and Aaa-rated consumer ABS (31 bps). Such an attractive spread pick-up in a sector that benefits from Agency backing is surely worth grabbing.   Treasury Valuation Chart 11Treasury Fair Value Models The current reading from our 2-factor Treasury model (based on Global PMI and dollar sentiment) pegs fair value for the 10-year Treasury yield at 2.94% (Chart 11). Our 3-factor version of the model (not shown), which also incorporates the Global Economic Policy Uncertainty Index, places fair value at 2.92%. PMIs across the world continue to surge. December PMI data show increases in the four largest economic blocs (U.S., Eurozone, China, Japan), and more broadly show that 86% of the 36 countries with available data currently have PMIs above the 50 boom/bust line. Meanwhile, bullish sentiment toward the U.S. dollar continues to trend lower in response to strong growth in the rest of the world (bottom panel). This is also a bearish development for U.S. bonds. For further details on our Treasury models please refer to U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com. At the time of publication the 10-year Treasury yield was 2.48%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com Alex Wang, Research Analyst alexw@bcaresearch.com Jeremie Peloso, Research Assistant jeremiep@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "How Much Higher For Yields?", dated October 31, 2017, available at usbs.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, "Ill Placed Trust?", dated December 19, 2017, available at usbs.bcaresearch.com 3 Please see U.S. Bond Strategy Weekly Report, "Proactive, Reactive Or Right?", dated December 12, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Proactive, Reactive Or Right?", dated December 12, 2017, available at usbs.bcaresearch.com 5 Please see U.S. Bond Strategy Weekly Report, "Ill Placed Trust?", dated December 19, 2017, available at usbs.bcaresearch.com 6 For further details on the model 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 Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights U.S. Treasuries: U.S. Treasury yields are too low relative to the strength of global economic growth and the rising trend in U.S. inflation expectations. Maintain below-benchmark duration exposure in the U.S., stay underweight Treasuries versus global bond benchmarks, and continue to favor TIPS over nominals. Canada: The Canadian economic data is moving from strength to strength, and now price and wage inflation data is moving higher. The Bank of Canada will hike rates next week with additional increases likely in 2018. Remain underweight Canadian government bonds and stay long inflation protection (both through linkers and CPI swaps). 2017 Model Portfolio Performance Wrap-Up: We closed the books on the first full calendar year of our model bond portfolio with a total return of 3.75%. This was a small -13bps of underperformance versus our custom benchmark, coming entirely from underweight positions on longer-dated developed market government bonds that offset the asset allocation gains from overweights to corporate debt. Feature Chart of the WeekGlobal Bond Yields Are Too Low 2018 has started much as 2017 ended, with growth-sensitive assets rallying alongside robust economic data. Most major global equity markets are already up 2-3% after the first week of the year, with the U.S. NASDAQ, Japanese Nikkei and Italian MIB indices advancing over 4%. Global credit markets are also off to a strong start, with spreads for U.S. High-Yield corporate debt and EM hard currency corporate debt tighter by -17bps and -8bps, respectively. Even commodity markets have joined the party, with the benchmark Brent oil price hitting the highest level in nearly three years. The pro-growth, pro-risk backdrop is keeping upward pressure on global government bond yields. This is occurring primarily through the inflation expectations component of yields, which are rising in all developed economies (even Japan). Real yields, which are not rising despite the strength of the broad-based global growth upturn (Chart of the Week), have been drifting lower, providing some offset to rising inflation expectations. The primary trend for global yields remains upward, however - especially if growth remains solid and inflation expectations continue to push higher, giving central banks like the U.S. Federal Reserve the confidence to continue hiking interest rates. We continue to favor below-benchmark duration exposure, and overweight corporate bond allocations versus government debt, for global fixed income investors over the next 6-9 months. U.S. Treasuries: Still More Reasons To Sell Than Buy U.S. Treasury market participants have a lot to things to be nervous about at the moment. Likely future Fed rate hikes, the weakening U.S. dollar, rising oil prices, ongoing U.S. labor market strength, persistently booming economic growth, the never-ending equity bull market, the potential impact of the Trump fiscal stimulus, the Fed starting its balance sheet runoff - all factors that should force bond investors to expect yields to rise. Yet longer-dated Treasury yields continue to trade too low relative to the bond-bearish fundamentals. The current benchmark 10-year Treasury yield at 2.48% remains well below the fair value from our 2-factor regression model, which is now up to 2.94% (Chart 2). That valuation gap of 46bps is close to the widest levels seen in July 2016 and September 2017, which were both episodes that proved to be excellent entry points for bearish Treasury positions. The two inputs into our Treasury yield model are the global manufacturing PMI and bullish sentiment towards the U.S. dollar (USD). The PMI is included as an indicator of global growth and currently sits at 54.5 - the highest level in nearly seven years - led by strong readings in almost every major economy (Chart 3). This has been the primary driver of the fair value for the 10-year Treasury yield since global growth bottomed out and began to accelerate in mid-2016. Chart 210-Year Treasuries Are##BR##Overvalued On Our Model Chart 3Global Growth##BR##Is Booming Sentiment towards the USD is the second input to our Treasury model. It is included as a weakening greenback represents an easing of monetary conditions that could trigger a need for more Fed rate hikes that can push the Treasury curve higher from the short-end (and vice versa for a rallying USD). At the same time, a depreciating USD can drive U.S. inflation higher through higher costs of imported goods & services, which can raise bond yields through higher inflation expectations or greater Fed tightening expectations (again, the opposite holds true for a strengthening USD). Right now, both the strong PMI and weak sentiment towards the dollar are boosting the fair value of the 10-year Treasury yield. The fall in value of the greenback is particularly unusual, as it is flying in the face of widening interest rate differentials between the U.S. and the rest of the world (Chart 4, top panel). This is clearly a function of the fact that global growth is rapidly improving - especially in Europe - but very few central banks have yet to respond to that growth with interest rate hikes that match what the Fed has been delivering. So while actual interest rate differentials remain USD-supportive, expectations of some eventual tighter monetary policy outside the U.S. that could narrow those interest rate gaps are triggering speculative inflows into non-USD currencies. With the trade-weighted USD now 5% below levels of a year ago, this should lead to higher headline inflation in the U.S. in the next few months (middle panel). Combined with the continued strength in global oil prices, that means that the two biggest factors that weighed on realized U.S. inflation- the USD rally and oil price collapse of 2014/15 - are now both acting to boost inflation expectations (bottom panel). Throw in the growing body of evidence that a tight U.S. labor market that is putting gentle upward pressure on wage growth, and U.S. inflation expectations - which still remain 40-50bps below levels consistent with the Fed's inflation target - should continue to move higher in the next six months. Rising longer-term inflation expectations would typically result in bear-steepening pressures on the Treasury yield curve. That is not happening at the moment, however, with the 2-year/10-year Treasury curve still at a relatively flat 53bps at the time this report went to press. The flatness of the Treasury curve has worried investors, and even some Fed officials, given the well-known leading relationship between the yield curve and U.S. economic growth. It is too early to draw any conclusions between the shape of the curve and future U.S. economic growth, however, for several reasons: As mentioned above, inflation expectations are still well below levels consistent with the Fed's 2% inflation target on the PCE deflator (which translates to 2.5% on the CPI index used to price TIPS and CPI swaps). Both the European Central Bank (ECB) and Bank of Japan (BoJ) are still buying bonds through their asset purchase programs, although at a slower pace than previous years. This continues to depress local bond yields in Europe and Japan with spillover effects into the U.S. Treasury market - even as the Fed begins the slow runoff of Treasuries from its massive balance sheet. Data on mutual fund and ETF flows shows that there has been significant and sustained buying of bond funds by U.S. retail investors over the past couple of months. There has also been net selling of equity funds, however, suggesting that U.S. retail investors are rebalancing as the equity markets surge higher. Investor positioning in the U.S. Treasury market is very short at the moment, with the J.P. Morgan survey of "active" bond manager duration exposure at an all-time low and the net positioning on Treasury futures now slightly favoring shorts (Chart 5). It makes little sense to interpret a flattening Treasury curve as a signal that the bond market believes that the Fed was making a policy mistake if professional bond investors were running massive duration underweight positions that would benefit if bond yields rise. Chart 4Upside Pressure On U.S. Inflation##BR##From Oil & The USD Chart 5Big Duration Underweight##BR##Among U.S. Bond Managers All these factors muddy the economic signal provided by the Treasury curve at the moment. Nonetheless, we remain of the view that the Fed would not continue on its rate hiking path without U.S. inflation expectations moving sustainably back to levels consistent with the Fed's inflation target. In other words, the Treasury curve must bearishly steepen first through rising inflation expectations before bearishly flattening later through actual Fed rate hikes. The latter will dampen future U.S. growth expectations and eventually result in a cyclical peak in longer-dated Treasury yields, but from levels closer to 3% on the 10-year after inflation expectations "fully" normalize. Bottom Line: U.S. Treasury yields are too low relative to the strength of global economic growth and the rising trend in inflation expectations. Maintain below-benchmark duration exposure in the U.S., stay underweight Treasuries versus global bond benchmarks, and continue to favor TIPS over nominals. The Bank Of Canada Keeps On Playing Catch-Up The Canadian economic story continues to be the best within the developed world. The year-over-year growth rate for real GDP accelerated to over 3% late last year, primarily on the back of robust consumer spending (Chart 6). Even the lagging parts of the economy, like business investment and government spending, began to perk up last year. The momentum remained powerful at the end of 2017, with the unemployment rate in December hitting a 40-year low. The economic boom forced the Bank of Canada (BoC) to begin lifting interest rates last year, with two 25bp hikes occurring in July and September that unwound the easing from 2015. The rapid pace of growth has absorbed spare capacity much faster than the BoC originally projected. More hikes will be required if the current pace of growth is maintained, particularly with the BoC estimating that the neutral policy rate is around 3% and the current Overnight Rate is only at 1%. The Canadian consumer has been enjoying a powerful shopping spree. Real consumer spending growth is at 4% on a year-over-year basis - the highest level since early 2008 (Chart 7). This is led by a powerful surge in spending on consumer durables, where annual growth has surged to 10% (middle panel). Consumer confidence is booming and Canadian workers are enjoying the fastest pace of income growth since 2014 (bottom panel). Chart 6Robust Canadian Growth,##BR##Led By The Consumer Chart 7Canadian Consumers Are##BR##Confidently Spending Surprisingly, the powerful surge in consumer spending has occurred alongside some cooling of the overheated Canadian housing market. The growth rates of existing home sales and prices have both decelerated massively from the pace of the boom years in 2012-16 (Chart 8). The performance of house prices in the three biggest Canadian cities is now a mixed bag, with Vancouver prices reaccelerating, prices in Toronto decelerating and prices in Montreal growing only modestly (middle panel). Regulatory actions to limit the speculative buying of Canadian real estate by foreigners has helped dampen the surge in house prices in some markets. Although the bigger macro-prudential measures designed to tighten mortgage finance rules and reduce the amount of leverage in Canadian housing transactions has likely had a bigger effect. Canadian banks must now conduct stress tests to check if borrowers are able to pay off their mortgages if Canadian interest rates continue to rise. This represents a reduction in the marginal supply of riskier mortgage lending that will help restrain house price inflation in Canada's major cities. In addition, the supply of Canadian homes is growing with new home-building activity, both for single and multiple units, having picked up and overall residential investment growth now up nearly 5% on a year-over-year basis (bottom panel). With signs that the Canadian housing market has stopped rapidly inflating, the BoC can focus its interest rate policy on domestic growth and inflation considerations without worrying about pricking the housing bubble. On that front, the latest edition of the BoC's Business Outlook Survey, released yesterday, provided plenty of reasons to tighten monetary policy further. The overall survey indicator surged back to the peak seen last summer just before the BoC delivered its first rate hike (Chart 9). Capital spending intentions also rebounded back to the 2017 peaks, which bodes well for future gains in investment spending (second panel). Chart 8Canadian Housing Looking##BR##A Bit Less Frothy Chart 9BoC Business Outlook Survey Signaling##BR##Tightening Capacity Constraints The most interesting parts of the Business Outlook Survey were the capacity utilization measures. A greater share of companies were reporting labor shortages (third panel), with the highest percentage of firms reported difficulties in meeting unexpected increases in demand since 2007 (bottom panel). This suggests that the recent surge in employment, wage growth and price inflation are all sustainable. Headline and core CPI inflation are up to 2.1% and 1.8%, respectively, as of November. This is around the midpoint of the BoC's 1-3% target range (Chart 10). The Bank of Canada forecasts that CPI inflation will continue to rise and remain near 2% target in 2018, but all the risks are to the upside. The unemployment rate is now down to 5.7%, the lowest level since 1976 and well below the OECD's estimate of the NAIRU level at 6.5%. Average hourly earnings growth has surged in response, rising to just under 3% on a year-over-year basis since the trough in early 2017. The Phillips Curve appears to be alive and well in Canada. Canadian interest rate markets have already responded aggressively to the stronger growth and inflation data. Our interest rate discounters now show that the money markets are now expecting 61bps of BoC rate hikes over the next six months and 91bps over the next twelve months (Chart 11). With a 25bp hike at next week's BoC meeting now priced with almost full certainty, the current market pricing suggests at least one more hike will happen by June and nearly three more hikes by year-end. That would be even more hikes than we expect from the Fed in 2018, which is important for the Canadian dollar (CAD). The CAD has appreciated 16% since it bottomed out in early 2016, occurring alongside the rise in global oil prices over the same period (second panel). The price of Canada's Western Select grade of crude oil has lagged the move in other oil benchmarks massively over the past several months, due to a lack of pipeline capacity getting oil out of Alberta that has created a supply glut. This may limit the degree to which additional gains in global energy prices benefit the Canadian dollar from a terms-of-trade perspective. This will not prevent the BoC from delivering additional rate hikes, however - especially if that merely matches the 75bps of Fed rate hikes that the FOMC is projecting, and which we expect, over the rest of the year. In terms of investment strategy, the combination of robust Canadian economic growth and rising inflation pressures leads us to continue recommending an underweight stance on Canadian government bonds, as we have maintained since July 11, 2017. This week, we are introducing two new tactical trades that should benefit as Canadian inflation moves higher and the BoC tightens more aggressively in response (Chart 12): Chart 10The Canadian Phillips Curve Is Not Dead Chart 11The Market Now Expects A Lot From The BoC Chart 12Two New Tactical Trades In Canada Short the June 2018 Canada Bankers' Acceptance futures contact vs. the December 2018 contract (middle panel). The market is now discounting the likely maximum amount of tightening that the BoC can deliver by year-end, while there are only little more than two hikes priced by June. Assuming that the BoC hikes next week, that means that there is only one more hike expected by June. With three more BoC meetings scheduled between next week and June, that provides plenty of opportunities for hawkish surprises from the BoC before then. In other words, this trade is a way to play for the BoC being forced to front-load more rate hikes into the first half of 2018 versus the latter half. Long 10yr inflation expectations through linkers versus nominal government bonds, or using CPI swaps (bottom panel). Given the pickup in domestic inflation pressures currently underway, plus the rise in global inflation coming from the surge in commodity prices, there is room for Canadian market-based inflation expectations to rise from the current level of 1.7%. Bottom Line: The Canadian economic data is moving from strength to strength, and now price and wage inflation data is moving higher. The Bank of Canada will likely hike rates next week with additional increases likely in 2018. Remain underweight Canadian government bonds. 2017 GFIS Model Bond Portfolio Performance: A Brief Review The turn of the year marked the end of the first full calendar year for the Global Fixed Income Strategy (GFIS) model bond portfolio. This now allows us to report the performance of the portfolio on the same basis as our clients. In the future, we will publish quarterly reviews of the portfolio returns after the end of each quarter in a calendar year (in April, July, October and January). The GFIS model portfolio returned 3.45% in 2017. This underperformed our custom performance benchmark (a blend of the Barclays Global Aggregate Index with global high-yield corporate debt) by -13bps (Chart 13). That underperformance can be entirely attributed to our government bond duration allocations, which lagged the benchmark by -46bps. Our recommended credit positions were a positive contributor, generating 33bps of outperformance primarily through overweights to U.S. Investment Grade and High-Yield corporate bonds. The detailed breakdown of the 2017 returns is presented in Table 1. In terms of the government bond portion of the portfolio, the underperformance can be isolated completely to the longest maturity bucket (10+ years). The combined performance of that bucket for all countries lagged that of the benchmark by -52bps. Given our expectation that global yield curves would bear-steepen in the latter half of 2017, it is no surprise that the bulk of our underperformance came by having too little exposure at the long-end. Also, having too much exposure in Japanese government bonds offering no yield also represented a major drag on the income component of the model portfolio's returns (Chart 14). Chart 13GFIS Model Bond Portfolio##BR##2017 Return Breakdown Table 1GFIS Model Bond Portfolio##BR##2017 Return Breakdown In terms of our credit allocations, favoring U.S. corporate exposure vs. non-U.S. corporates was the right call, generally speaking (Chart 15). However, we did not have enough portfolio weight in that trade to offset the drag on the overall yield from the Japan government bond overweight. Chart 14GFIS Model Portfolio Government Bond Performance Attribution By Country Chart 15GFIS Model Portfolio Spread Product Performance Attribution Looking ahead, the new model bond portfolio allocation for 2018 that we discussed in our final report of 2017 should offer a better chance of outperforming the benchmark.1 Specifically, we dialed down the Japan overweight, increased the U.S. Investment Grade corporate bond overweight, and reduced the curve steepening exposure in Euro Area governments. This not only boosted the overall yield of the portfolio, but also moderated the overall portfolio duration underweight. This portfolio will do well in the first half of 2018 if our base case of an inflation-driven rise in global government bond yields, led primarily by the U.S. where corporate debt is also expected to outperform Treasuries, comes to fruition. Bottom Line: We closed the books on the first full calendar year of our model bond portfolio with a total return of 3.75%. This was a small -13bps underperformance of versus our custom benchmark, coming entirely from underweight positions on longer-dated developed market government bonds that offset the asset allocation gains from overweights to corporate debt. Robert Robis, Senior Vice President Global Fixed Income Strategy rrobis@bcaresearch.com Ray Park, Research Analyst ray@bcaresearch.com 1 Please see BCA Global Fixed Income Strategy Weekly Report, "Our Model Bond Allocation In 2018: A Tale Of Two Halves", dated December 19th 2017, available at gfis.bcaresearch.com. Recommendations The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Highlights Upbeat economic reports for December set the stage for a solid 2018. The FOMC minutes acknowledged the flatter curve and only a minority of members discounted the signal from the curve. A majority thought that a tighter labor market would lead to higher inflation. The Citi Economic Surprise Index is peaking, but risk assets should hold up as the Index rolls over. Feature The first week of 2018 brought more good news for risk assets. U.S. stocks beat bonds, oil prices rose, and credit spreads narrowed amid a solid set of economic data. Several high-profile U.S. companies announced share buybacks, and/or one-time bonuses or wage increases linked to the tax cut plan passed by Congress at the end of 2017. Moreover, there were hints of further economic stimulus as lawmakers from both sides of the aisle discussed relaxing the sequester rules that would lift federal spending this year. Markets shrugged off a fresh round of saber rattling between the U.S. and North Korea. Gold prices nudged higher and the U.S. dollar fell despite the upbeat economic news. December's reports on manufacturing and service sector ISM, vehicle sales and the labor market, along with November's numbers on construction spending, trade and factory orders, all lifted estimates for Q4 GDP and boosted the prospects for corporate earnings in Q4 2017 and beyond. Chart 1 shows that the elevated ISM figures provide a favorable backdrop for earnings and sales in 2018. Moreover, Chart 2 indicates that IP, a proxy for S&P 500 sales, is poised to advance in 2018 and provide a lift to corporate profits. We will preview the S&P 500's Q4 2017 earnings reports in next week's U.S. Investment Strategy. Chart 1Favorable Macro Backdrop For Earnings And Sales Chart 2ISM Components Suggest IP Poised To Accelerate The Atlanta Fed GDP Now estimate stood at 2.7% on January 5, while the New York Fed's Nowcast for Q4 GDP was a healthy 4% (Chart 3). Both soundings are well above the FOMC's assessment of the economy's long-term potential growth rate (1.8%) and puts GDP growth in 2017 above the Fed's forecast. The implication is that the output gap pushed deeper into positive territory as 2017 ended, setting the stage for higher inflation in 2018. The December 2017 jobs report, released last Friday, January 5, does not change BCA's outlook for the U.S. economy or the Fed. The U.S. economy added a lower than expected 148,000 new jobs in December, which left the unemployment rate unchanged at 4.1%. Despite the softer than anticipated data, the 3-month average of payrolls growth is still a very healthy 204,000. The monthly increase in wages quickened to 0.3% m/m in December, up from 0.1% m/m last month. However, annual wage inflation remains modest at just 2.5% (Chart 4). Chart 3U.S. Economic Growth Well##BR##Ahead Of Potential In Q4 Chart 4Labor Market Still Tightening Despite##BR##Soft December Report The indications for Q4 GDP growth are solid. Aggregate hours worked rose 2.5% at an annualized rate in Q4 2017. Assuming modest growth in productivity, the payrolls data are consistent with over 3% GDP growth in Q4. There is nothing in the December payroll data to suggest that the underlying trajectory in the U.S. economy has changed. The economy continues to grow above trend. Wage gains are modest at the moment, but should accelerate as the labor market keeps tightening with above-trend GDP growth. This upbeat economic outlook is also supported the December 2017 non-manufacturing ISM survey, also released last Friday. While the overall index fell from 57.4 to 55.9, it is still consistent with solid GDP growth. Moreover, the employment index rose from 55.3 to 56.3, which signals firm job gains, and the prices paid index held steady at a fairly elevated level of 60.8. Bottom Line: It's been solid start to 2018 and it's steady as she goes for the U.S. economy and the Fed. FOMC Minutes: A Rubric BCA's U.S. Bond Strategy service expects that the 2/10 yield curve will languish between 0 and 50 bps in 2018. The curve will steepen from 51 bps at the end of 2017 through mid-year 2018, and then flatten into year-end (Chart 5). Which asset classes would benefit if our curve call is accurate? BCA's "The Bucket List"1 explains our view of the curve in 2018 and details the past performance of various U.S. assets in differing yield curve environments. Chart 5A Flat Yield Curve Is OK For Most Risk Assets BCA expects that the yield curve will first steepen in 2018, then become flatter, ultimately spending most of the year between 0 and 50 bps. A flat curve is the ideal environment for the S&P 500 and the stock-to-bond ratio. However, small cap stocks struggle when the curve is flat; BCA's view is that small caps will outperform large caps in 2018. A flat yield curve raises the risk of a sell-off in high yield, but provides a favorable grounding for oil, which is in line with BCA's fundamental view. BCA expects EPS growth will be positive this year; earnings growth is higher 75% of the time when the curve is flat. The yield curve's slope was a focus of debate at the FOMC's December 12-13, 2017 meeting. Participants cited several reasons for the flat curve2: recent increases in the target range for the federal funds rate; reductions in investors' estimates of the longer-run, neutral real interest rate; lower longer-term inflation expectations; lower term premiums Fed economists recently updated their quantitative assessments of the FOMC's minutes. The note provides a guide (Table 1 in the Fed paper3 and Tables 1 and 2 below) to the number of quantitative descriptors in the minutes (one, a couple, a few, etc.). We use this rubric to assess the committee's latest views on the yield curve and inflation. Table 1FOMC Assessment Of The Yield Curve Table 2FOMC Assessment Of Inflation In short, the FOMC acknowledged the flatter curve and only a minority of members discounted the signal from the curve. Moreover, a majority thought that a tighter labor market would lead to higher inflation. Only one participant held the view that secular trends were muting inflation. Bottom Line: BCA expects the Fed to deliver 3 to 4 rate hikes in 2018, which is still not fully priced in by the market. Investors should maintain below-benchmark duration in fixed income portfolios. Asset allocators should remain overweight stocks versus bonds. Growth is strong and the yield curve is not inverted yet. Therefore, it is still early to de-risk portfolios. Is Economic Surprise Peaking? The Citigroup (Citi) Economic Surprise Index is elevated relative to its recent history, but it may have further to run. Economic prospects were cheery following the 2016 presidential election and the economic data exceeded those lofty projections, aided by a warmer than usual winter. However, the temperate conditions borrowed activity from the spring, which was cooler and wetter than normal, and the combination of lofty expectations and seasonal distortions sent the Citi Economic Surprise Index spiraling lower through mid-year 2017. Since its bottom in June 2017 at -78.6%, the index climbed for 135 days before its peak in late December 2017 (Chart 6, panel 1). On average since 2010, the Citi Index moved from trough-to-peak in 96 days, which means the recent run-up was much longer than usual. However, that phenomenon may have been due to the raised economic expectations and variable weather patterns at the start of 2017. Chart 6Economic Surprise Index Has Surged, But Expectations Remain Muted At 80.7%, the Index has been above zero for 68 days (Chart 6, panel 1). It typically takes 46 days for it to climb from zero to its zenith. Table 3 shows the performance of financial markets and other assets after the Index moves from zero to the peak. The most recent episode (October through December 2017) matched historical averages across most asset classes, although the underperformance of small caps versus large ran counter to the past as the Surprise Index climbed from zero. Table 3Risk Assets Perform Well As Surprise Index Climbs Since 2010, the Index has stayed above 40 for an average of 51 days (Chart 6, panel 1). The Index has been over 40 since November 16, 2017, or 35 days. This suggests that it can remain elevated for another month or so before it again moves lower. However, the Index is mean reverting and investors wonder what will happen to risk assets after economic surprise rolls over. Table 4 and Chart 7 shows the performance of key financial markets and commodities when the Citi Index returned to zero from 40-plus. There have been six such intervals since 2010. On average, gold and oil perform well as the surprise index dips to zero. Stocks and credit outperform Treasuries during these episodes, and small caps beat large caps. Rising economic surprise (Table 3) is a more favorable environment for stocks, credit and oil than when the surprise index is rolling over. However, the performance of gold and small caps is better after the Citi Surprise Index peaks (Table 4). Table 4Risk Assets Hold Up When Citi Surprise Index Rolls Over Chart 7U.S. Assets As Economic Surprise Rolls Over Nonetheless, muted economic expectations will limit the downside in the Index in the coming months. Panel 3 of Chart 6 shows that the outlook for both hard and soft economic data remained muted through the end of November 2017, especially when compared with the significant improvement in economic prospects in late 2016 and early 2017. Bottom Line: Risk assets outperformed as the Citi Economic Surprise Index climbed in the second half of 2017. The Index can stay near recent peaks for several more months thanks to subdued economic forecasts, but it will roll over eventually. However, the elevated level of the Index suggests that there are near-term risks for equities and credit because a lot of good economic news is already priced in. Still, we recommend that investors ride out the volatility given our view that stocks will outperform bonds in the next 6-12 months. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com 1 Please see BCA Research's U.S. Investment Strategy Weekly Report "The Bucket List", published December 18, 2017. Available at usis.bcaresearch.com. 2 https://www.federalreserve.gov/monetarypolicy/fomcminutes20171213.htm 3 https://www.federalreserve.gov/econres/notes/feds-notes/the-fomc-meeting-minutes-an-update-of-counting-words-20170803.htm

Highlights Should the U.S. 10-year T-bond yield approach 3% it would be a red flag, and a trigger to downgrade equities. Equity investors should stay overweight defensive-heavy Switzerland and Denmark. Contrary to what the consensus is expecting, global growth will lose steam in the first half of 2018. EUR/USD will continue to trend higher through 2018 as long-term interest rate differentials converge further. The multi-year prognosis for GBP/USD is higher. U.K. parliamentary arithmetic simply does not support a hard Brexit. Furthermore, a hard Brexit would require either a North/South or East/West hard border in Ireland, which will be politically impossible to deliver. Feature A happy and prosperous 2018 to you all! In this first report of the year, we describe some investment outcomes in 2017 that at first glance seemed odd or unexpected; but that on deeper reflection provide valuable insights for 2018. Some of these insights deviate substantially from the BCA house view. Bonds Became More Risky Than Equities The first oddity of 2017 concerns the 'drawdowns' suffered by bonds and equities. A drawdown is defined as an investment's peak to trough decline. In 2017, the odd thing was that the drawdowns suffered by government bonds - a supposedly safe asset-class - were equal to or worse than those suffered by equities - a supposedly risky asset-class (Chart of the Week, Chart I-2 and Chart I-3). Chart of the WeekBonds Suffered Worse Drawdowns Than Equities Chart I-2Bonds Suffered Worse Drawdowns Than Equities Chart I-3Bonds Suffered Worse Drawdowns Than Equities Contrary to classical theory, empirical evidence now proves that investors do not define an investment's risk in terms of its volatility, the fluctuations of its return around a mean. Instead, investors define risk as the ratio of large and sudden drawdowns versus potential gains. This unattractive asymmetry in an investment's return is technically known as negative skew. And it is as compensation for this negative skew that investors demand an excess return, the so-called 'risk premium'. Significantly, at low bond yields, the mathematics of bond returns necessarily means that their negative skew increases. The risk of large and sudden drawdowns rises while the prospect for price gains diminishes. But if bond risk becomes 'equity-like', it follows that equities' prospective long-term return should become 'bond-like'. Meaning, equities should no longer offer a meaningful risk premium over bonds. Is this the case? According to my colleague Martin Barnes, BCA Chief Economist, the answer appears to be yes - at least in certain major markets. In BCA's Outlook 2018, Martin projects that from current valuations U.S. equities are set to deliver a total nominal return of 2.6% a year to 2028 - almost indistinguishable from the 2.5% a year that a U.S. 10-year T-bond will deliver over the same period. But the mathematics of bond pricing tells us that the negative skew on bond returns fully disappears when a yield approaches 3%. At which point the risk of bonds once again declines to become 'bond-like', and the required return on equities should once again rise to become 'equity-like'. This higher required return would necessarily require today's equity prices to drop, perhaps substantially. Admittedly in Europe there is a bigger gap between the expected returns from equities and bonds than there is in the U.S. The trouble is that global capital markets move together and a chain is only as strong as its weakest link. Hence, one lesson for 2018 is that investors should downgrade equities to neutral should the U.S. 10-year T-bond yield approach 3%. In this event, investors should redeploy the funds into U.S. T-bonds, because any substantial adjustment in risk-asset prices would trigger supportive flows into haven bonds, reversing the spike in yields. Euro/Dollar Hit A 3-Year High EUR/USD ended 2017 touching 1.21, a 3-year high. At first glance, this might seem odd given that the ECB has committed to maintaining its zero and negative interest rate policy for at least another year while the Federal Reserve has already hiked interest rates five times. But EUR/USD is not tracking short-term rate differentials. It is tracking long-term rate differentials, and EUR/USD at a 3-year high is fully consistent with the 30-year T-bond/German bund yield spread converging to its narrowest for several years (Chart I-4). Chart I-4Further Convergence In Long-Term Interest Rate Differentials Will Support EUR/USD Where will this yield spread go from here? Let's consider both sides of the spread. On the ECB side, policy is at the realistic limit of ultra-looseness, so policy rate expectations cannot go significantly lower, but they can go higher. On the Federal Reserve side, long-term policy rate expectations are not far from our upper bound of the 'high 2s' at which risk-assets become vulnerable to a sell-off, perhaps substantial. So these interest rate expectations cannot go sustainably higher, but they can go lower. Considering this strong asymmetry, the most likely outcome is that the 30-year T-bond/German bund yield spread will continue to converge. The upshot is that EUR/USD will continue to trend higher through 2018. No Connection Between Economic Outperformance And Stock Market Outperformance Chart I-5The Eurostoxx50 Underperformed Even Though##br## The Euro Area Economy Outperformed 2017 proved that there is no positive correlation between relative economic performance and relative equity market performance. For example, the euro area was one of the best performing developed economies, yet the Eurostoxx50 was one of the worst performing stock market indexes (Chart I-5). This seems odd, until you realise that major stock market indexes are dominated by multinational rather than domestic stocks. And that when stock markets have vastly different sector weightings, the sector effect completely swamps the domestic economy effect. Therefore the first decision for international equity investors should never be which regions to own. The first decision should always be which sectors to own, and above all whether to tilt to cyclicals or defensives. The regional and country allocation then just drops out automatically. At the moment, our mini-cycle framework for global growth suggests tilting to defensives rather than to cyclicals. Global growth experiences remarkably consistent - and therefore predictable - 'mini-cycles', with half-cycle lengths averaging 8 months. As the current mini-upswing started last May we can infer that it is likely to end at some point in early 2018 (Chart I-6 and Chart I-7). So one surprise could be that global growth will lose steam in the first half of 2018 rather than in the second half - contrary to what the consensus is expecting. Chart I-6The Current Mini-Upswing##br## Is Long In The Tooth Chart I-7China Has Driven The Global 6-Month##br## Credit Impulse Higher We will provide further ammunition for our mini-cycle thesis in next week's report. In the meantime, we will leave you with one ramification of paring back equity exposure to cyclicals and redeploying to defensives. Stay overweight defensive-heavy Switzerland and Denmark. Realpolitik Will Prevent A Hard Brexit For the FTSE100, the paradox is that its relative performance is negatively correlated with relative economic performance. When the U.K. economy outperforms, the FTSE100 underperforms. And vice-versa (Chart I-8). Chart I-8FTSE 100 Relative Performance Is The Inverse ##br##Of U.K. Economic Relative Performance The simple explanation is that FTSE100 multinational sales and profits tend to be denominated in dollars and euros, whereas the FTSE100 index is denominated in pounds. The upshot is that an outperforming U.K. economy weighs on the U.K. stock market because a strengthening pound diminishes the FTSE100's multi-currency profits in pound terms. And vice-versa. Compared to a year ago, investors can be more optimistic about the long-term prospects for the U.K. economy and the pound (and therefore expect long-term underperformance from the FTSE100). This is because after the unexpectedly disastrous 2017 election for Theresa May, the parliamentary arithmetic simply does not support a hard Brexit. Furthermore, a hard Brexit would require either a North/South or East/West hard border in Ireland, which will be politically impossible to deliver. The constraints that come from this realpolitik means that Brexit's endpoint will retain much of the current trading relationship with the EU, albeit the journey to that eventual destination is likely to be a wild roller coaster ride. Therefore, the multi-year prognosis for GBP/USD is higher. But investors who want to optimize their timing into 'cable' can wait for one of the inevitable roller coaster dips in 2018. Dhaval Joshi, Senior Vice President Chief European Investment Strategist dhaval@bcaresearch.com Fractal Trading Model* We are delighted to say that three of our recent trades quickly hit their profit targets: short bitcoin 29%, long silver 4.5% and long NZD/USD 3%. Against this, short Nikkei/long Eurostoxx50 hit its 3% stop-loss. This week's trade recommendation is to go short palladium. Set a profit target of 6% with a symmetrical stop-loss. This leaves us with three open trades. Chart I-9 For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields Interest Rate Chart II-5Indicators To Watch##br## - Interest Rate Expectations Chart II-6Indicators To Watch##br## - Interest Rate Expectations Chart II-7Indicators To Watch##br## - Interest Rate Expectations Chart II-8Indicators To Watch##br## - Interest Rate Expectations