Consumer Discretionary

Overweight After reaching their post-GFC highs in the middle of 2017, the S&P homebuilders index put in a seven-year bottom in October, though has since recovered to roughly the level where we upgraded to overweight.1 The fall came as homebuilder sentiment gave up some of its optimism (second panel), driven largely by higher mortgage rates and the resulting lower affordability (third panel). However, we believe the market priced in faltering homebuilder optimism well before the industry itself. Anecdotally, Toll Brothers (TOL, notably not a member of the S&P homebuilders index but a major homebuilder nonetheless) recently delivered quarterly results where new orders fell for the first time in four years, including a precipitous 39% decline in California, their biggest market, and offered guidance below analyst expectations. This glum news was met by a market that took the stock higher that day. Our inference is that negativity is fully priced in to this index and with the overall housing market index squarely above the 50 boom/bust line (second panel) and earnings growth expectations coming back to reality (bottom panel), the S&P homebuilders index is set to continue its recovery. We reiterate our overweight recommendation. The ticker symbols for the stocks in this index are: BLBG: S5HOME - DHI, LEN, PHM.   1      Please see BCA U.S. Equity Strategy Weekly Report, “Indurated” dated September 24, 2018, available at uses.bcaresearch.com.

Special Report

Feature The European stock market has a hidden gem: its clothing and accessories sector. Since the turn of the millennium, the sector’s profits are up by a thousand percent (Feature Chart). In this Special Report we propose that the megatrend has further to run, as its principle driver is still very much in place. Consumption patterns are becoming more female. Feature ChartEuropean Clothes Profits Are Up A Thousand Percent! One of Europe’s major, and largely neglected, success stories is the dramatic rise in the percentage of the working-age population in employment. This major success story stems from another success story: the structural and broad-based increase in the female labour participation rate – which has surged from 57 percent in 1995 to 68 percent today (Chart I-2-Chart I-4). Yet the story is far from over.1 Chart I-2European Male Labour Participation Is Flat...   Chart I-3...But European Female Labour Participation Is Surging   Chart I-4...So The Percentage Of The European Population In Work Is Surging Why Job Creation Favours Women Two things are driving the megatrend in female participation. One is a paradoxical feature of the current technological revolution. As we explained in The Superstar Economy: Part 2, Artificial Intelligence (AI) excels at tasks that we perceive as difficult: those requiring the application of complex algorithms and pattern recognition to a narrowly defined goal, such as making a highly-engineered product or managing a stock portfolio. This poses a big threat to jobs in manufacturing and finance, employment sectors which happen to be male-dominated.2 Conversely, AI still struggles at tasks that we perceive as easy: those requiring adaptable movements, or reading and responding to people’s emotions and intentions. If you are good at controlling a disruptive class of 7-year olds, or calming a nervous patient before giving him an injection, your human skills are still in big demand. But education, healthcare, and social care – the employment sectors that are creating the most jobs – employ three times as many women as men. With AI still in its infancy, the established pattern of job destruction and creation will continue to favour women over men (Table I-1). Table I-1AI Is A Greater Threat To Men The other driver of the megatrend in female participation is a raft of European legislation designed to make work more family friendly: flexible working time, generous paid maternity and paternity leave, and subsidised childcare (Table I-2-Table I-4). Sharing the responsibility of childcare between mothers, fathers and external helpers has allowed tens of millions of European women to enter and remain in the labour force. Table I-2Generous Maternity Pay In Europe And Japan   Table I-3Improving Paternity Pay In Europe And Japan   Table I-4Affordable Childcare In Europe And Japan Nevertheless, the megatrend has a lot further to run. For the ultimate end-point, look at the Scandinavian countries which started legislating such policies in the early 1970s, around twenty years before the rest of Europe. As a result, in Sweden, labour force participation rates for women and men have now converged to almost identical: 81 versus 84 percent (Chart I-5). Chart I-5In Sweden, Labour Force Participation For Women And Men Is Almost Identical The combination of the two drivers – employment growth favouring female-dominated sectors and employment becoming more family friendly – means that net job creation in Europe will be mostly due to more women joining the workforce. An important consequence is that consumption patterns will continue to become more female. But what does that mean? How Women’s Spending Differs From Men’s Spending In the main spending categories of housing, food and healthcare, women and men tend to show near-identical spending behaviours. But there are three sub-categories where there are significant differences. Men considerably outspend women on vehicle purchases: cars account for around 8 percent of disposable income for men versus 4 percent for women. Against this, women spend more on personal care products and services: 2 percent versus 0.5 percent. This is the reason behind our long-standing successful overweight recommendation in the European personal products sector which we maintain (Chart I-6). However, the sub-category in which women outspend men by even more is clothes and accessories: estimates average around 6.5 percent for women versus 2.5 percent for men.3 Chart I-6Personal Product Profits Set To Grow Very Strongly It follows that as consumption patterns become more female, we should expect to see a steady rise in spending on clothes and accessories as a share of total consumer spending. Has this been the case? In the U.K. – where the data is easily available – the answer is yes (Chart I-7). Having said that, other factors are also at play. A generalised deflation in clothes prices (Chart I-8) is also generating a strong tailwind to sales volumes (rather than values). More about this later. Chart I-7More Real Spending On Clothes... Chart I-8Partly Because Clothes Prices Are Falling... Of course, the more compelling evidence is the thousand percent growth in the European clothes sector’s profits since the turn of the millennium. However, with the sector dominated by top brands such as LVMH and Hermes, could a more plausible explanation come from strong economic growth, until recently, in the emerging markets such as China? The answer is yes to the extent that many of the emerging economies are experiencing the same structural uptrends in female participation, and this supports our investment thesis. Still, this cannot be the main driver, because in recent years the connection between the fortunes of the emerging economies and the European clothes sector has been weak (Chart I-9).   Chart I-9The Connection Between Emerging Markets And European Clothes Is Weak There is another obvious question: is the market already aware of, and fully priced for, the megatrend? We think not, as most investors we meet are surprised by the structural uptrend in female participation, the on-going dynamics behind it, and the implications for consumer spending patterns. Understandably, the European clothes sector does trade at a valuation premium to the market (Chart I-10). But for many companies, the recent market hiccup has pulled down their valuation premiums to close to, or below, the long-term average from which the price has previously outperformed very strongly. Chart I-10The Valuation Premium On European Clothes Is Close To Its Long-Term Average What Is In The Clothes Basket? Pulling all of this together, the companies in our European clothes and accessories basket need to meet several criteria: A dominant or significant exposure to women’s clothes and/or accessories. A top-end brand (or brands) giving the company pricing power, and mitigating the very strong deflation in clothes prices. Avoid ‘fast fashion’. A reputation for sustainable development. A track-record of profit growth during the past decade. A forward price to earnings (PE) multiple of less than 25. A market capitalisation of at least €5 billion. On the basis of these criteria, our European clothes and accessories basket contains four names: LVMH, Kering, Luxottica, and Burberry (Table I-5). Hermes meets most of the criteria but, trading on a forward PE close to 35 is very richly valued. Table I-5The European Clothes Basket To be clear, this is not a short-term trade. Investors who buy the clothes basket outright need to have a multi-year investment horizon. Those investors who must also protect short-term performance should instead overweight the clothes basket relative to the broad market. Dhaval Joshi, Senior Vice President Chief European Investment Strategist dhaval@bcaresearch.com Footnotes 1 Please see the European Investment Strategy Special Report “Female Participation: Another Mega-Trend” published on April 6, 2017 and available at eis.bcaresearch.com 2 Please see the European Investment Strategy Special Report, “The Superstar Economy: Part 2”, January 19, 2017 available at eis.bcaresearch.com. 3 Source: Bureau of Labor Statistics Consumer Expenditure Survey 2016 via SmartAsset, and Paymentsense.

Underweight (High-Conviction) While the probability of a housing recession remains low, we are concerned that too much euphoria is already priced in the S&P home improvement retail (HIR) index, and there are high odds that next year HIR will suffer the same fate as homebuilders did this year (top panel). Thus, we are downgrading the S&P HIR index to underweight and adding it to the high-conviction underweight list for 2019. Fixed residential investment (FRI) as a percentage of GDP is up 50% from trough to the recent peak, whereas relative HIR performance is up 170% in the same time frame. Our worry is that optimistic sell side analysts’ relative profit forecasts will be hard to attain, let alone surpass as FRI is steadily sinking (second panel). Worrisomely, our HIR model has plunged on the back of the wholesale liquidation in lumber prices and rising interest rates (bottom panel). Lumber deflation will prove a profit headwind as building supply Big Box retailers make a set margin on wood products. Bottom Line: Rich valuations will be tough to maintain amidst weak FRI, lower lumber prices and higher interest rates. We downgraded to an underweight position on Monday and added the S&P HIR index to our high-conviction underweight list; please see Monday’s Weekly Report for more details. The ticker symbols for the stocks in this index are: BLBG: S5HOMI - HD, LOW.

Highlights Portfolio Strategy Higher interest rates, with the Federal Reserve tightening monetary policy three more times in the next seven months, will be the dominant theme next year. All four of our high-conviction underweight calls are levered to this theme. The later stages of the U.S. capex upcycle underpin three of our high-conviction overweight calls for 2019. Recent Changes Downgrade the S&P Home Improvement Retail index to underweight today. Trim the S&P Interactive Media & Services index to a below benchmark allocation today.  Table 1 Feature Fed policy will dominate markets next year as the dual tightening backdrop – rising fed funds rate and accelerated downsizing of the Fed balance sheet – remains intact. Two weeks ago we raised the question: is the Fed tightening monetary policy too far too fast?1 In more detail, we put the latest monetary tightening cycle in historical perspective and examined trough-to-peak moves in the fed funds rate since the 1950s (Chart 1). Chart 1Too Far Too Fast? A good friend I call “the smartest man in California” correctly pointed out that 500bps of tightening today is not the same as in the 1970s or 1980s. Chart 2 adjusts for that by including the average nominal GDP growth rate during these tightening episodes and adds more color to each era. As a reminder, the latest cycle that commenced in December 2015 is already 25bps above the median, if one uses the Wu-Xia shadow fed funds rate to capture the full quantitative easing effect, and above-average nominal output growth. Chart 2Trough-To-Peak Tightening Cycle Already Above Historical Median Trying to answer the question, we are concerned that as the Fed remains committed to tighten monetary policy three more times by mid-2019, a yield curve inversion looms, especially if the U.S. economy suffers a soft patch in the first half of next year (please refer to our Economic Impulse Indicator analysis in the October 22ndand November 19th Weekly Reports). This would signal at least a pause, if not reversal, in Fed policy. With that in mind, this week we are revealing our high-conviction calls for 2019. Four of our calls are a play on this tightening monetary backdrop that is one of BCA’s themes for next year.2 The later stages of the U.S. capex upcycle underpin three of our high-conviction calls. Table 22018 High-Conviction Calls Recap However, before we highlight our 2019 high-conviction calls in detail, Table 2 tallies our calls from last year. We had a stellar performance in our 2018 high-conviction calls with an average excess return of 11.6% versus the S&P 500. As the year turns the corner, closing out the remaining calls brings down the average relative return to 7.5%, still a very impressive number, with a total of ten hits and only two misses for the year.    Anastasios Avgeriou, Vice President U.S. Equity Strategy anastasios@bcaresearch.com     Software (Overweight, Capex Theme) Software stocks are our first hold out from last year’s high-conviction overweight list, levered to the capex upcycle theme. Chart 3 shows that relative capital outlays and the share price ratio are joined at the hip. Software upgrades offer the simplest, quickest and most effective capital deployment, especially when productivity gains ground to a halt. Importantly, leading indicators of overall capex remain upbeat and should continue to underpin software profits. Beyond capex, M&A has been fueling software stock prices. It did not take long for the large CA acquisition to get surpassed by RHT and more recently SYMC was also rumored to be in play (Chart 3). Inter-industry M&A activity is reaching fever pitch and this frenzy is bidding up premia to stratospheric levels. The push to the cloud, SaaS and even AI has boosted the appeal of software stocks and brought them to the forefront of potential takeout candidates. These are secular trends and will likely continue to gain steam irrespective of the different stages in the business cycle. As a result, software stocks should remain core tech holdings in equity portfolios. The recovery in the software price deflator (Chart 3), a proxy for industry pricing power, corroborates the upbeat demand backdrop. With regard to financial statements, software stocks have pristine balance sheets with more cash on hand than debt, which sustains the net debt-to-EBITDA ratio in negative territory. Interest coverage is great at 10x and free cash flow generation is expanding smartly. The ticker symbols for the stocks in this index are: BLBG: S5SOFT - MSFT, ORCL, ADBE, CRM, INTU, RHT, ADSK, SNPS, CTXS, ANSS, CDNS, FTNT and SYMC. Chart 3Software   Air Freight & Logistics (Overweight, Capex Theme) Air freight & logistics stocks are the second hold out from our high-conviction overweight list, although we added it to list only in late-March. This transportation sub-index laggered is a capex and trade de-escalation play for the first half of 2019. Importantly, energy costs comprise a large chunk of freight services input costs and the recent drubbing in oil markets will boost margins especially on the eve of the busiest season for courier delivery services (top panel, Chart 4). On that front, there are high odds that this holiday sales season will be another record setting one, as wage inflation is underpinning discretionary incomes. Keep in mind that the accelerating domestic manufacturing shipments-to-inventories ratio confirms that demand for hauling services is upbeat. The implication is that rising demand for freight services will buoy industry profits and lift valuations out of their recent funk (Chart 4). Firming industry operating metrics also tell a positive story and suggest that relative share prices will soon take off. Air freight pricing power has been healthy, in expansionary territory and above overall inflation measures. While the U.S./China trade tussle and the appreciating greenback are clear risks to our sanguine S&P air freight & logistics transportation subindex, most of the grim news is already reflected in depressed relative forward profit estimates, bombed out valuations and washed out technicals (Chart 4). The ticker symbols for the stocks in this index are: BLBG: S5AIRF - FDX, UPS, EXPD and CHRW. Chart 4Air Freight & Logistics   Defense (Overweight, Capex Theme) We have been overweight the pure-play BCA defense index since late-2015 and there are high odds that this juggernaut that really commenced with the George Walker Bush presidency remains in a secular growth trajectory. Our strategy is to add exposure on any meaningful pullbacks and keep this index as a structural overweight within the GICS1 S&P industrials index. The recent drawdown offers such an opportunity and we are adding this index to the 2019 high-conviction overweight list. The rise of global "multipolarity" - or competition between the world's great nations - and the decline of globalization, along with a global arms race and increased risk of cyber-attacks, have been documented in our "Brothers In Arms" Special Report. These trends all signal that global defense related spending will remain upbeat in the coming decade.3 In the U.S. in particular, where military spending in absolute terms is greater than the rest of the world put together, defense spending and investment have bottomed and will continue to accelerate (Chart 5). In fact, the CBO continues to project that defense outlays will jump further next year. While such a breakneck pace is clearly unsustainable, President Trump is serious about upgrading and updating the U.S. military in order to keep China's geopolitical and military ascendancy in check (as well as to deal with Russia and Iran).4 The upshot is that defense outlays will continue to expand into the 2020s. Such a buoyant demand backdrop is music to the ears of defense contractor CEOs, and represents a boost to defense equity revenue growth prospects. This capital goods sub-industry has extremely high fixed costs and thus any increase in top line growth flows straight to the bottom line. Put differently, defense contractors enjoy high operating leverage. No wonder M&A activity is robust: at least four large deals have been announced in the past year that are underpinning takeout premia. A closer look at operating metrics corroborates that defense goods manufacturers are firing on all cylinders. New orders recently jumped to fresh all-time highs and the industry's shipments-to-inventories ratio is rising, on track to surpass the 2008 peak. Unfilled orders are also running at a high rate, signaling that factories will keep on humming at least for the next few quarters. Importantly, the industry is not standing still and is making significant investments. U.S. defense capex as reported in the financial statements of constituent firms is growing at roughly 20%/annum or twice as fast as overall capex (Chart 5 on page 7). While interest coverage has been modestly deteriorating, it is twice as high as the overall market (Chart 5 on page 7). Impressively, defense ROE is running near 30%, again roughly double the rate of the broad market. The ticker symbols for the stocks in the BCA defense index are: LMT, LLL, NOC, GD and RTN. Chart 5Defense   Consumer Discretionary (Underweight, Higher Fed Funds Rate Theme) We recommend investors avoid the consumer discretionary sector that suffers when interest rates rise. Chart 6 depicts this inverse correlation consumer discretionary equities have with interest rates, especially the fed funds rate. Most discretionary equites are levered off of floating rates and thus any increase in the fed funds rates gets reflected immediately in banks' prime lending rate. Also, most consumer debt is floating rate debt and thus tighter monetary conditions, at the margin, dampen consumer debt uptake and, as a knock-on effect, weigh on discretionary consumer outlays. Recently we highlighted that, now that the Fed has been raising rates and allowing bonds to roll off its balance sheet, volatility is making a comeback. Unsurprisingly, the consumer discretionary share price ratio is inversely correlated with the VIX index, signaling that more pain lies ahead for this early cyclical index (VIX shown inverted, Chart 6). Sentiment and technical indicators also point to more downside ahead for this interest-rate sensitive index. Our sector advance/decline line is waning and EPS breadth has plunged. Worrisomely, sell-side analysts are penciling in an extremely optimistic 5-year outlook with EPS growth 23.4%/annum or 1.4 times higher than the overall market. Clearly this is not realistic as it assumes a tripling of EPS in the coming 5 years. Relative EPS estimates have already given way as AMZN commands very little EPS weight, despite its massive market cap weight (30% of the S&P consumer discretionary sector), and suggests that relative share prices will converge lower (Chart 6 on page 9). As a result, the 12-month forward P/E ratio is trading at a 24% premium to the broad market and significantly above the historical mean. Technicals are almost as extended as relative valuations and cyclical momentum has likely peaked, warning that a downdraft in relative share prices looms (Chart 6 on page 9). Chart 6Consumer Discretionary   Home Improvement Retail (Underweight, Higher Fed Funds Rate Theme) While the probablity of a housing recession remains low, we are concerned that too much euphoria is already priced in the S&P home improvement retail (HIR) index, and there are high odds that next year HIR will suffer the same fate as homebuilders did this year (Chart 7). Thus, we are downgrading the S&P HIR index to underweight and adding it to the high-conviction underweight list for 2019. Fixed residential investment (FRI) as a percentage of GDP is up 50% from trough to the recent peak, whereas relative HIR performance is up 170% in the same time frame. Our worry is that optimistic sell side analysts' relative profit forecasts will be hard to attain, let alone surpass as FRI is steadily sinking (Chart 7). Worrisomely, our HIR model has plunged on the back of the wholesale liquidation in lumber prices and rising interest rates (Chart 7). Lumber deflation will prove a profit headwind as building supply Big Box retailers make a set margin on wood products. Select industry operating metrics suggest that the easy profits are behind HIR. Not only is our productivity growth proxy (sales per employee) on the verge of deflating, but also an inventory surge has sunk the HIR sales-to-inventories ratio into the contraction zone. Finally, there is rising supply of new and existing homes for sale already on the market, and that puts off remodeling activity at least until this supply glut clears (months' supply shown inverted, Chart 7). The ticker symbols for the stocks in this index are: BLBG: S5HOMI - HD, LOW. Chart 7Home Improvement Retail   Short Small Caps/Long Large Caps (Higher Fed Funds Rate Theme) The days in the sun are over for small cap stocks and we are compelled to put the size bias favoring large caps in our high-conviction calls list for 2019. Small caps are severely debt saddled. Sustained small cap balance sheet degradation is worrying, with S&P 600 net debt-to-EBITDA close to 4 compared with less than 2 for the SPX (Chart 8). Such gearing is fraught with danger as the default rate has nowhere to go but higher. Small and medium enterprises (SMEs) have a higher dependency on bank credit as opposed to the bond market access that mega caps enjoy. Most bank credit is floating rate debt and so are lines of credit, and as the Fed remains firm on tightening monetary policy, interest expense costs are skyrocketing for SMEs. In a relative sense this will weigh on net profits. Moreover, small caps are a lot more sensitive to interest rates, and the selloff in the 10-year Treasury note heralds more pain in 2019 (Chart 8). Small caps are high(er) beta stocks and when volatility spikes they underperform large caps. When the Fed ballooned its balance sheet and dropped the fed funds rate to zero it suppressed volatility. Now that the Fed has been decreasing the size of its balance sheet and raising interest rates, this is working in reverse and volatility is making a comeback as we have been highlighting in our research, and will continue to weigh on small caps (VIX shown inverted, middle panel, Chart 8). Another way to showcase small caps' riskier status is the close correlation they have with the relative EM equity share price ratio. When EMs outperform the SPX, small caps follow suit and vice versa. Importantly a wide gap has opened recently and we suspect that it will narrow via small caps following the EM higher beta stocks lower (SPX vs. EM ratio shown inverted, fourth panel, Chart 8 on page 12). Chart 8Small Vs. Large   Interactive Media & Services (Underweight, Higher Fed Funds Rate Theme) In our initiation of coverage on the S&P interactive media & services index,5 we highlighted three key risks that offset the revenue & profit growth vigor of this group, comprised almost entirely of Alphabet (Google) and Facebook. These were a renewed regulatory focus, rapid unpredictable changes in tastes & technology and an appreciating U.S. dollar. It is the first of these that has risen most dramatically since that report. Tack on the inverse correlation these growth stocks have with interest rates (top panel, Chart 9) and that is causing us to lower our recommendation to underweight and include this index in the high-conviction underweight list for 2019. Increasing regulatory efforts on technology will be a key theme next year, one we explored this past summer.6 Our conclusion was that both antitrust (particularly in the case of Alphabet) and privacy regulation (particularly in the case of Facebook) added significant risk to these near monopolies; calls for legislating both have dramatically amplified. Tim Cook, Apple’s CEO, recently commented that more regulation for Facebook and Alphabet was inevitable; we agree. While the form such regulation might take remains open to debate (for example, the U.S. could adopt an EU-style General Data Protection Regulation (GDPR)), we fear the associated headline risk (not to mention likely profit headwinds) will impair stock prices in the S&P interactive media & services index. This communication services sub-index is particularly prone to such a risk when it already trades at close to a 40% valuation premium to the broad market (middle panel, Chart 9 on page 14). Adding insult to injury is the PEG ratio that is trading at a 60% premium to the broad market (bottom panel, Chart 9 on page 14). In the face of the Fed’s sustained tightening cycle these extreme growth stocks are vulnerable to massive gravitational pull. The ticker symbols in the stocks in this index are: S5INMS – GOOGL, GOOG, FB, TWTR and TRIP. Chart 9Interactive Media & Services Footnotes 1      Please see BCA U.S. Equity Strategy Report, "Manic Market," dated November 19, 2018, available at uses.bcaresearch.com. 2      Please see BCA The Bank Credit Analyst Report, "OUTLOOK 2019: Late-Cycle Turbulence", dated November 26, 2018, available at bca.bcaresearch.com. 3      Please see BCA U.S. Equity Strategy Special Report, "Brothers In Arms," dated October 31, 2016, available at uses.bcaresearch.com. 4      Please see BCA Geopolitical Strategy Special Report, "A Global Show Of Force?" dated October 10, 2018, available at gps.bcaresearch.com. 5      Please see BCA U.S. Equity Strategy Special Report, "New Lines Of Communication," dated October 1, 2018, available at uses.bcaresearch.com. 6      Please see BCA U.S. Equity Strategy Special Report, "Is The Stock Rally Long In The FAANG?", dated August 1, 2018, available at uses.bcaresearch.com.   Current Recommendations Current Trades Size And Style Views Favor value over growth Favor large over small caps

  Underweight The S&P restaurants index has had an exceptional month, following surprisingly healthy results from both McDonalds and Starbucks, which collectively represent approximately 80% of the index. We think the move is short-sighted and we would be sellers into the strength. The positive results at these restauranteurs would typically be reflected in outsized forward earnings growth estimates. However, this is not the case; as shown in the second panel, estimates for the S&P restaurants index are falling behind the broad market. Given the index’s rise without a corresponding earnings lift, the valuation multiple has soared and is now at a level 50% higher than the market (third panel). Meanwhile, the index has been dining out on an unhealthy diet of debt and balance sheets are stretched to extreme levels (bottom panel). In the absence of an unlikely surge of cash flow, particularly given the headwinds an appreciating U.S. dollar represents, a painful cycle of belt-tightening lies ahead. Bottom Line: Tepid earnings growth, high valuations and bulging leverage are a recipe for stock price declines in the S&P restaurants index; stay underweight. The ticker symbols for the stocks in this index are: BLBG: S5REST - MCD, SBUX, YUM, DRI, CMG.

Feature Asset Allocation Amid Late-Cycle Turbulence BCA today published its 2019 Outlook, Late-Cycle Turbulence,1 our annual discussion with long-time clients Mr X and his daughter, Ms X. Recommendations This note is simply to update the Global Asset Allocation (GAA) service’s recommended portfolio in the light of our expectations for the next 12 months, as detailed in the 2019 Outlook. The table on the right summarizes our key views, with the full details of recommended weights for a standard portfolio on page 3 of our PDF. Our overall view of markets has not changed significantly. We remain concerned about late-cycle pressures that will ultimately undermine asset prices. However, it is hard to put precise timing on when the U.S. economy will peak. For the next few quarters, it will continue to benefit from loose fiscal policy and still accommodative monetary policy. Although global growth is slowing, a recession is not on the immediate horizon. Therefore, we think it is premature to move underweight on equities. A further sell-off might even present a buying opportunity. Accordingly, we remain neutral on global equities, with a preference for the U.S., underweight global bonds, and overweight cash. The only significant change we make to GAA’s recommended portfolio is to reduce Japanese equities to neutral. Profit margins are at record highs but are vulnerable to narrowing as wage growth picks up and sales to emerging markets come under further pressure. Our model suggests that Japanese profit growth will turn negative in year-on-year terms by the second half of 2019. Barring evidence of more reflationary stimulus from China, Japan will be challenged to outperform global equities. We have used the funds taken from Japan to increase our overweight in U.S. equities. We make two minor changes to our equity sector recommendations, lowering Industrials to underweight and raising Consumer Discretionary to neutral. This is in line with our generally defensive sector tilts. We expect to see a further slowdown in China and, in the U.S. too, capex growth has probably peaked for this cycle. These factors should be negative for capital goods producers. Consumer Discretionary should benefit from a late-cycle rise in wages in developed economies, and contains some structural growth companies in the online commerce space, which are now more attractively valued after the recent sell-off. We will publish our next Monthly Portfolio Update on January 2.   Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com   Footnote 1 Please see BCA Outlook titled “Late-Cycle Turbulence” dated November 26, 2018 available at bca.bcaresearch.com Recommended Asset Allocation

Special Report

Highlights U.S. housing's immediate past will not repeat, ... : It is understandable that investors who lived through the financial crisis are acutely sensitive to any sign of trouble in housing, but none of the factors that helped precipitate the crisis are in place now. ... and its older history will only rhyme: Home construction slowdowns have a good record of signaling recessions, but residential investment's steadily waning share of GDP has chipped away at its influence. The current housing soft patch is not over, but it's unlikely to get much worse, ... : The rapid rise in mortgage rates sharply reduced affordability, but it still remains at a very comfortable level relative to history. Inventories of new and existing homes are very low, and the pace of new construction continues to run slightly behind household formation. Most importantly for the expansion, there were no construction excesses in this cycle that need to be worked off. ... so we don't think it's sending any broader signal about the economy: A tiny contraction in residential investment is not a harbinger of recession, nor is it an indication that monetary policy is already tight. Feature Desynchronization has been the name of the game in 2018. The U.S. economy, already ahead of its peers in putting the crisis in its rear-view mirror, has gotten an additional fillip from the fiscal stimulus package. Global growth, on the other hand, has been slipping. As Fed chair Jay Powell put it last week, the rest of the world is "gradual[ly] chipping away" at the U.S., but there "is not a terrible slowdown" in the global ex-U.S. economy. Global conditions have not slowed enough to get the Fed to interrupt its tightening campaign, but signs of softness outside of the U.S.'s borders have been popping up like mushrooms after the rain. With disappointments having been few and far between in the U.S., any pockets of weakness that do appear attract immediate attention. Against this backdrop, the slowing in housing - residential investment has now contracted for three consecutive quarters - is making some investors a little uneasy. We have spent a good deal of time within BCA debating housing's recent softness, its outlook, and its implications for financial assets and the economy, and clients are increasingly inquiring about our views. Housing's Recent Past Housing is top of mind for many investors because it was at the center of the financial crisis. Residential mortgages were ground zero of the credit bubble that systemically threatened the banking system. Wobbles in housing bring back unpleasant memories of the searing trauma that unfolded just ten years ago. With the dot-com mania and the financial crisis having occurred just a decade apart, the financial media, and many strategists, analysts and investors are on high alert for the next crash. The concerns are understandable, but conditions today are nearly the polar opposite of conditions in 2005 and 2006. There is nothing even remotely bubble-like about the current housing market. The critical weakness back then was the shunning of time-tested underwriting standards, as revealed by the homeownership rate. An average of just over 64% of households owned their own homes for the first three decades of the ownership series in a remarkably steady pattern,1 but a steady debauching of standards pushed the rate to above 69% at its peak (Chart 1, top panel). Chart 1Too-Easy Lending Standards ... The homeownership rate was built on a foundation of increasingly unserviceable mortgages (Chart 1, bottom panel). Prices surged (Chart 2, top panel), flippers flooded the market, and homebuilders ramped up production to meet the ensuing demand (Chart 2, second panel). When the music stopped, the housing market was left with unprecedentedly large inventories of unsold homes (Chart 2, third panel); the banking system's primary source of collateral was poised to suffer a body blow; and a hiring surge that played out over a decade and a half was unwound in just two years (Chart 2, bottom panel). Chart 2... Made Housing Unstable Housing In The Current Cycle Current conditions are much more stable. The homeownership rate is back to its time-tested levels. New housing supply has generally undershot the smoothed trend in household formations ever since the crisis ended (Chart 3, top panel). Inventories are strikingly low when adjusted for the overall size of the housing stock (Chart 3, middle panel). The vacancy rate is low (Chart 3, bottom panel), and there is no construction employment cliff. Most importantly from a stability perspective, the Basel III/Dodd-Frank regulatory framework makes it very difficult to replicate the reckless credit conditions that enabled the housing bubble. This cycle has been devoid of housing excesses. Chart 3Plenty Of Room For More Homes A broader historical context reveals that housing has been exerting steadily less influence on the economy across the entire postwar era. We have a good deal of sympathy for the argument that the postwar business cycle has been a consumption cycle, largely led by housing,2 but it's possible that the crisis marked housing's last hurrah as a driver of recessions. Residential investment's share of GDP exploded when pent-up demand was released upon the return of servicemen and women needing homes for their burgeoning families (Chart 4). The construction of the interstate system, and the network of subsidiary roads that sprang up to connect to it, facilitated the creation of the suburbs, and Levittown-style tract housing communities had to be built from scratch to meet the demand. Chart 4The Incredible Shrinking Impact Of Housing Activity The baby boom kept demand for more, and larger, houses going strong. Once grown themselves, the baby boomers helped keep household formation growth flush. The baby boomers are now net sellers, however, and will be at an increasing rate across the next couple of decades. The time trend of residential investment's share of GDP is stark, and demographics are poised to keep it going as long as the baby boomers are divesting their holdings. The bottom line is that we do not think housing is the business cycle this time around. It is a highly cyclical part of the economy, and its fluctuations will still be felt, but its influence on the overall economy has been steadily waning for 70 years, and it is not currently in a position to exert a powerful drag. It would be overstating matters to say that housing booms cause recessions, but they've been observed at the scene of the crime in every recession of the last 60 years except for the dot-com bust. In this cycle, the barely visible white area above the trend line in Chart 4 is nowhere near large enough to give rise to a big swing below the trend line, and inspire a patch of gray shading on its own. The ratio of housing starts to the existing stock of homes (Chart 5) reinforces the message of residential investment's declining contribution to overall output. The United States has been augmenting and/or replacing the existing stock of homes at a steadily diminishing rate for 60 years. Assuming that the rate of obsolescence has remained roughly constant, it seems that there has simply been less to build once the suburban frontier was settled. Even against the declining time trend, however, residential construction activity in this cycle has not revived enough to require a correction. Chart 5Tinkering Around The Edges We attribute the current softness to the backup in mortgage rates over the last twelve months. 100 basis points may not seem like the end of the world, but the rise in interest rates has been sudden, and it is entirely plausible to think that it has sent some marginal first-time buyers to the sidelines. The Housing Affordability Index is way below its 2013 peak, but remains quite high relative to its pre-ZIRP history (Chart 6, top panel). The sudden drop in the index has been a function of mortgage payments (Chart 6, second panel) as sudden moves almost always are - the median home price (Chart 6, third panel) and the median income series (Chart 6, bottom panel) are much less variable. Chart 6Mortgage Rates Drive Affordability We expect that rates will go still higher, but our bond strategists don't think it will happen any time soon. They see rates consolidating for a while as the economy digests the sharp move higher, and favorable year-over-year comparisons cool off inflation's upward momentum over the coming months. Our above-consensus view on the terminal fed funds rate is not housing friendly. Housing will have to contend with ongoing bond-market headwinds, but we don't expect another move of this magnitude will recur in such a concentrated time frame. Bottom Line: Housing may face a headwind from higher rates for at least another year, but a big drop-off in activity is not in the cards. There are no current cycle excesses that need to be unwound, and housing has become too small a part of the economy to induce a recession on its own. Housing Demand And The Fed Funds Rate Cycle The notion that mortgage rates are to blame for the housing soft patch raises some questions about our assessment of the monetary policy backdrop. Is it possible that a funds rate that's proximally related to a slowdown in housing demand is not impacting consumer demand for other goods or services, or corporate demand? Could there be multiple equilibrium fed funds rates? If not, is the housing soft patch a sign that the economy is actually in Phase II of the cycle, and not Phase I? We are unperturbed by the three-quarter contraction in residential investment, which one has to squint to see (Chart 7). We do not believe that housing demand has reached an inflection point; we simply think that prospective monthly mortgage payments have moved so fast that some buyers have temporarily stepped aside. Given that buying a home still looks quite inviting by the historical standards of the affordability index, conditions are not yet restrictive. Ex-the ZIRP era, the index had not exceeded 140 for more than three decades (Chart 6, top panel). If homes are still affordable relative to history, then housing would seem to support our equilibrium fed funds rate model's assessment that monetary policy remains accommodative. Chart 7Not Much Of A Downturn Yet We view the state of policy as binary for the economy as a whole, even if some activity is necessarily more rate-sensitive. While some marginal investment projects cease to generate positive prospective net present value any time interest rates rise, encouraging or discouraging activity is a universal condition. The broader investment-relevant question is whether or not our assessment that the fed funds rate cycle has not yet transited from Phase I to Phase II is correct (Chart 8). If the economy is still in Phase I, and will remain there for a year, our constructive take on the economy and financial markets still applies. If it's shifted to Phase II, however, the empirical record says investors should be paring back risk. Chart 8The Fed Funds Rate Cycle The preponderance of evidence supports the idea that we remain in Phase I. Real-time measures of activity remain robust. Credit performance remains very good, so banks are still eager lenders. Employment is surging, and a follow-up dose of fiscal stimulus in 2019 should keep all the plates spinning for another year. As macro investors, and students of cycles, we are as eager as anyone to recognize the inflection point as swiftly as possible, but the data series we follow do not indicate that it is approaching. We continue to abide by our equilibrium fed funds rate model's benign conclusion. Investment Implications Although housing's direct impact on GDP has steadily waned, it remains an important part of the economy, given how it feeds into several other elements of consumer demand. Three consecutive quarters of contraction in residential investment are worthy of notice, but such a run has occurred before without provoking a recession, and the contraction to date has been awfully modest in any event. We do not view the slowdown as the beginning of the end for the expansion. We also do not view it as a sign that monetary policy is tighter than we originally judged. We expect that the ongoing surprise over the rest of this cycle will be that the neutral fed funds rate is considerably higher than the market consensus expects. We therefore think that investors should continue to maintain benchmark exposure to risk assets while remaining underweight Treasuries and holding all bond exposure below benchmark duration. Since we think the expansion remains in place, supported by accommodative monetary policy, we view the recurring mini-scares provoked by data points like housing's soft patch as potential opportunities to put our cash overweight to work. Doug Peta, Senior Vice President U.S. Investment Strategy dougp@bcaresearch.com 1 Over the 120 quarters through the end of 1994, a mean 64.3% of households owned a home, with a standard deviation of 0.6%. Only 22% of the quarterly observations were more than a standard deviation away from the mean, as opposed to the 32% predicted by the normal distribution. 2 Leamer, Edward E., "Housing IS the Business Cycle," NBER Working Paper No. 13428, September 2007. http://www.nber.org/papers/w13428

Special Report

Highlights Chinese pro-consumption policy stimulus will likely stabilize Chinese household consumption growth at 8-10% over the next 12-15 months, with service consumption continuing to be the key driver. Our research shows that Chinese nominal retail sales of consumer goods are currently growing at only 3-4%, significantly lower than the 9-10% pace that the Chinese government has reported, and that the market has commonly quoted. We expect it to rebound moderately to 4-6% in 2019 on the government's pro-consumption stimulus. The services sector including healthcare, education, travel, entertainment, sports, high-tech, daycare for kids, nursing homes for the elderly, and so on will likely experience strong growth. In the consumer discretionary space, car sales will also likely rebound as the country may soon release stimulus measures supporting the auto industry. For now, we advise overweighting consumer discretionary stocks versus the benchmark. We also recommend going long consumer discretionary versus consumer staples. Feature The Chinese economy is in transition from investment- and export-led growth to consumer-led growth. With faltering investment growth and escalating China-U.S. trade tensions, the strength and durability of Chinese household consumption has become all the more important to the country's economic growth. To address increasing challenges facing the economy, the government over the past several months has released a slew of policies aimed at stimulating domestic consumption. Our focus in this week's report is to outline these policies and in turn gauge what the strength of Chinese household consumption will be over the next 12-15 months. In order to do this, some key questions need to be addressed, including: What is the current growth rate of household consumption? What pro-consumption policies have already been implemented, and what additional policies are likely on the way? How effective will stimulus be on Chinese household consumption this time around? One of our key findings is that Chinese nominal retail sales of consumer goods - a common proxy for Chinese household spending - is currently growing at only 3-4%, significantly lower than the 9-10% pace the Chinese government has reported, and that the market has commonly quoted (Chart 1). Chart Retail Sales Growth Measure: Which One Is Accurate? Another important finding is that recent pro-consumption policy stimulus will likely increase household income levels by 400-500 billion RMB. In addition, we expect more pro-consumption policies from the Chinese government later this year or early 2019 - i.e., cutting car sales taxes or giving out subsidies to encourage households' purchases of automobiles, especially in rural areas, and/or lowering the policy rate to spur spending by reducing households' borrowing costs. This will stabilize Chinese household consumption growth at 8-10% over the next 12-15 months, with service consumption continuing to be the key driver. Making Sense Of The Data According to the National Bureau of Statistics (NBS) data, Chinese nominal household consumption accounted for about 40% of GDP last year, and grew 8.2% year-on-year (about 9-10% in 2015 and 2016). We estimate that currently about 65-70% of Chinese household consumption is consumer goods, with the remainder going to services. Goods consumption Chinese retail sales figures are probably the most-often-used among market participants as a proxy for Chinese household consumption, despite the fact that the data only provide a partial picture of Chinese household spending: spending on consumer goods. Based on the NBS's definition, Chinese total retail sales of consumer goods refer to the sum of retail sales of commodities sold to urban and rural households for household consumption, and to social institutions for public consumption for non-production purposes. Chinese total retail sales also include online goods sales but do not include online service sales. They also do not include many service sectors including education, medical care, travel, entertainment, eldercare and childcare. In short, while Chinese retail sales cannot represent the full picture of Chinese household consumption, they can indeed reveal the strength of Chinese household consumption on consumer goods. The most quoted retail sales growth data by the majority of market participants is from the NBS - a straight growth number that the bureau reported every month - which recently decelerated to 9% (the dotted line in Chart 1). The bureau does not give out information about how to calculate this growth data. The NBS also reports the level data of retail sales every month, from which the year-on-year growth actually plunged to 3.8% (the solid line in Chart 1). Which one is more accurate? All of the findings below suggest the validity of the growth estimates we calculated from the level of NBS retail sales. For major consuming discretionary goods like cars, washing machines, air conditioners, refrigerators and TVs, all products excluding TVs exhibited a sharp drop in sales volume growth this year (Chart 2). Chart 2Falling Sales Volume Nearly Across The Board From Discretionary Goods... Some major consumer staples such as dairy products, soft drinks and liquor - also experienced a sharp decline in sales-volume terms (Chart 3). Chart 3...To Major Consumer Staples The sub-categories of total nominal retail sales in value terms also showed a significant slowdown in terms of urban and rural, and in terms of commodity goods and catering (Chart 4). Chart 4Weakness In Retail Sales From Urban To Rural Meanwhile, 26 out of 31 provinces experienced retail sales growth slower than 6% for the first six months, with three provinces - Shandong, Jilin and Guizhou - in contraction. Why did Chinese retail sales experience such a significant drop this year? We believe it is because households' sentiment and willingness to consume has diminished considerably (Chart 5). Chart 5Falling Marginal Propensity To Consume The cracking down of peer-to-peer lending, falling stock prices and high mortgage payments this year have all reduced household wealth. Mortgage interest payments currently account for nearly 50% of the nation's household disposable income, higher than 45% a year ago.1 In addition, rising China-U.S. trade tensions have also increased uncertainty on future income growth and affected confidence. Service consumption If our estimate of Chinese retail sales growth can correctly capture the strength of consumer goods consumption, what data can be used to measure services consumption. Chart 6 can at least provide some sense in gauging the strength of household service consumption, as tourism, medical services and entertainment services (i.e., movie box office receipts) are all major household service consumption components. In the meantime, online services sales can also somewhat reflect the overall strength of Chinese household services consumption. Chart 6Services Consumption Still Growing At A Double Digit Pace Chart 6 clearly shows that despite the growth deceleration, nominal services consumption growth is currently still quite strong - in the range of 10-15% - considerably higher than the 3-4% growth in nominal consumer goods consumption. To gauge how Chinese nominal household consumption growth will be going forward, we need to assess the pro-consumption policies that have already been implemented. Consumption Stimulus A flurry of pro-consumption policies has been announced over the past several months, aiming at spurring consumer spending to support the country's underlying economic growth. Personal tax cuts and tax exemptions will increase households' ability to spend, while improvements in the quality of goods and services supplied and more availability of high-quality products will also encourage consumption. On October 21, China unveiled a new income tax law to boost consumption. The law increases the tax-free threshold from 3,000 RMB per month to 5,000 RMB per month and expands the lower tax brackets, effective October 1, 2018. It also adds new itemized tax deductions related to education, housing, eldercare, childcare and medical care, which will come into effect on January 1, 2019. Additional details of the new itemized deductions have so far not been released. The Ministry of Finance estimates that the tax changes will collectively lift household incomes by approximately 320 billion RMB. This is equivalent to about 1% of household consumption expenditures, or about 0.4% of GDP. Given that the total amount of personal income tax was 1.2 trillion RMB last year, the total tax deduction from the new income law and new itemized tax deductions should be much smaller than the amount of total personal income tax. Assuming 40-50% of the 1.2 trillion RMB personal tax will be deducted in 2019, this will be equal to about 500-600 billion RMB in household incomes (1.6-1.9% of household consumption expenditures, and about 0.6-0.7% of GDP). On September 20, the government released a policy guideline: "New measures to spur residential consumption." Two weeks later, on October 11, the government announced a "three-year (2018-2020) action plan to stimulate domestic consumption." The government's plan is geared to facilitating a virtuous cycle in which boosting consumption leads to supply innovation, and subsequently improvement in new consumption growth. According to the plan, the authority will widen the openness of seven key service sectors for private and / or overseas companies to enter in sectors such as tourism, culture, sports, healthcare, eldercare, home services and education/training. The country aims to develop rental markets, promote new-energy automobiles, support high-tech products (VR, robots, etc.), encourage green consumption and upgrade the quality of existing goods and services. Insufficient high-quality supply in these service sectors have in the past curbed consumption growth to some extent. By boosting the supply of high-quality services, the government expects to increase consumption in these sectors. Starting on July 1, China reduced import tariffs on 1449 imported items, resulting in a decline in average import tariffs from 15.7% to 6.9%. Starting on November 1, the government further lowered tariffs with most-favored nations on an additional 1585 items with the average tax rate falling from 10.5% to 7.8%. Clearly, there are two trends from these policies. First, the services sector including healthcare, education, travel, entertainment, sports, high-tech, childcare, eldercare, and so on will benefit most, as households in general have high demand for these services and are willing to spend more on these sectors (Chart 7). Chart 7Service Consumption Vs. Consumer Staples Consumption: Higher Growth For example, while China's aging population will have increasing demands for medical and eldercare service, the termination of the one-child policy will continue to boost demand for childcare and education services. Food and clothing accounts for about 35% of total Chinese household consumption expenditures (Chart 8), significantly higher than the 21% proportion in South Korea. Meanwhile, Chinese consumers spend 11% of their disposable income on education, culture and recreation, lower than the 17% figure in South Korea. Chart 8Chinese Household Consumption Structure Second, the supply of high-quality consumer goods and high-quality services will strongly increase in response to rising demand of wealthier Chinese consumers. This increase in supply will be met by both domestic production of goods and services and overseas imports. What additional policies could be implemented in the remainder of 2018 and 2019? The government may release more supportive policies to promote car sales - i.e., reducing the sales tax on cars with a capacity of 1.6L or lower, or providing subsidies on car purchases. They have implemented similar stimulus measures since 2008. If recent pro-consumption policies and supportive policies for the auto industry still cannot revive household consumption strongly enough, the authorities may cut the policy rate to spur additional spending. After knowing the probable scale of the pro-consumption stimulus, we can now put everything together to see what Chinese household consumption growth could be in 2019. How Strong Will Household Consumption Be? Structurally, we believe growth in Chinese household consumption is facing strong headwinds, including lower household income growth in real terms (inflation-adjusted) because of slowing productivity growth and rising household debt levels (Chart 9). Chart 9Structural Headwinds For Chinese Household Consumption Growth However, over the next 12-15 months, we still expect the government's pro-consumption policies to be able to stabilize domestic household consumption growth at 8-10%. We estimated in the first section that the new income law and itemized tax deduction policy will likely release about 500-600 billion RMB of income available for spending. The ratio of marginal propensity to consume gauges the proportion of one additional unit of disposable income spent on consumption. We estimated that the marginal propensity to consume for Chinese households is currently at about 50%. This will result in 250-300 billion RMB spending on household consumption, equaling about 0.7-0.8% of 2017 Chinese retail sales of consumer goods (36.6 trillion RMB), or 0.8-0.9% of household consumption expenditures. Autos will be another major potential driver of overall household consumption growth. China has stimulated the car industry by slashing the auto sales tax from 10% to 5% in 2009-2010 and again in 2015-2016. As a result, the volume of passenger car sales jumped 50% in 2009 and 15% in 2016, respectively (Chart 10). While car sales have dropped each time the stimulus measures have expired, a temporary growth rebound in auto sales in 2019 is still possible. Chart 10The Government May Stimulate The Auto Market Again As car sales volumes are currently in double-digit contraction, the Chinese government is likely to implement similar stimulus measures in late 2018 or early 2019. If so, Chinese car sales in volume terms may rebound by 5-10% in 2019. By the end of last year, the measure of urban households with cars was about 37.5 out of 100. There is still plenty of upside, with the rural areas having much bigger potential for car sales than urban areas. The value of Chinese auto sales was 4.2 trillion RMB last year. It increased 280 billion RMB in 2016 and 220 billion in 2017, but decreased 220 billion for the first nine months of this year. Assuming a 5% growth in the auto sales value next year because of the stimulus, it will be about 200 billion RMB increase, equivalent to 0.2% of 2017 GDP or 0.6% of household consumption expenditures. Although households have already taken out much more in the way of consumer loans for purchases of homes and other day-to-day expenses, with plenty of consumption-related stimulus policy in place, consumer loan growth will likely continue to grow in the double digits in 2019 (Chart 11). In September, household loans for short-term consumption (non-mortgage) grew at 28% year-on-year. Chart 11Consumer Loan Growth May Remain Strong In 2019 Chart 12 shows that the breakdown of household borrowing - medium- and long-term consumption loans (mostly mortgage loans) accounted for 60% of total household borrowing. Chart 12Most Of Consumption Loans Are Mortgage With the property market now slowing down and a gradual decline in the Chinese central bank's PSL lending,2 property sold has been decelerating (Chart 13). This may lead to less mortgage borrowing, leaving more loans available for short-term spending. Chart 13Household Borrowing In 2019: Less For Mortgage And More For Consumption? How different is this round of stimulus versus the previous two episodes? First, the strength of household consumption growth due to recent policy stimuli will be much weaker than the 2009-2010 and 2015-2016 episodes (Chart 14). Chart 14Stimulus Impact On Household Consumption Growth In 2019: Less Than Previous Episodes Home appliance markets like TVs, air conditioners, washing machines and refrigerators have already entered a mature phase. On average, as of the end of 2017 there were already 133 TVs, 100.3 air conditioners, 97.1 washing machines and 97.2 refrigerators for every 100 urban households (Chart 15, top panel). Even in rural areas, as of the end of last year there were 120 TVs, 52.6 air conditioners, 86.3 washing machines and 91.7 refrigerators for every 100 households, significantly higher than 2008 levels (Chart 15, bottom panel). Chart 15Home Appliance Markets: More Mature Than The Auto Market Second, this time the stimulus is focusing more on the services sector, while the previous two episodes were more on consumer goods. As result, this time the stimulus will have much less impact on commodities than the previous two episodes, in which major commodity goods sales and production experienced significant growth. Overall, China's pro-consumption policies will likely stabilize Chinese household consumption growth at 8-10% over the next 12-15 months, with services consumption remaining the key driver. We expect household service consumption to continue to grow at 10-15%, and retail sales growth to rebound to 4-6% from 3-4%. Investment Implications Chinese pro-consumption policy will likely benefit services and the automobile industry more than consumer staples. Meanwhile, commodity sectors may not benefit much. For now, we recommend overweighting the domestic consumer discretionary sector versus the Chinese CSI300 benchmark, a trade that we are initiating as of today (Chart 16, top panel). The sector's relative P/E and P/B valuations versus the benchmark also suggest its relative attractiveness (Chart 16, middle and bottom panels). Chart 16Overweight Consumer Discretionary Versus Benchmark China's pro-consumption stimulus also warrants the opposite position of what was one of our most successful trades over the past year (long investable staples / short investable discretionary), which we closed at the end of September for a profit of 48%.3 Within the domestic market, investors should go long consumer discretionary versus consumer staples, a trade that we are also initiating as of today (Chart 17). In addition to the cyclical tailwind from policy, relative valuation ratios suggest that the former is likely to outperform the later. Chart 17Go Long Consumer Discretionary Versus Consumer Staples Finally, our conclusion that policymakers are likely to succeed at stabilizing household consumption growth has implications beyond the relative performance of consumer stocks. Our outlook for a stable consumer over the coming year supports the argument that China will not push for a significant reacceleration in credit growth as a response to ongoing economic weakness, and argues in favor of our view that the "strike price" of the China put option has fallen. As we have noted in previous reports, we have no doubt that Chinese policymakers will eventually move to a maximum reflationary stance if they feel that the existing slowdown will lead to deep, threatening economic instability. But it will be impossible for investors and policymakers to make a judgement about the true odds of such an outcome without hard evidence of the magnitude of the tariff-induced export shock, which for now remains obscured by trade front-running (which may persist until the new year). This means that it is too soon to bottom fish deeply oversold Chinese financial assets (such as A-shares), and China-related plays more generally. Ellen JingYuan He, Associate Vice President Emerging Markets Strategy EllenJ@bcaresearch.com 1 Pease refer to Table 1 in the China Investment Strategy Special Reports "China's Property Market: Where Will It Go From Here? ", dated September 13, and Table 1 in the Emerging Markets Strategy Special Report "China Real Estate: A Never-Bursting Bubble?", dated April 6, 2018, available at ems.bcaresearch.com. 2 Pease see China Investment Strategy Special Report "China's Property Market: Where Will It Go From Here?", dated September 13, 2018, available at cis.bcaresearch.com. 3 Pease see China Investment Strategy Special Reports "GICS Sector Changes: The Implications For China", dated September 26, 2018, available at cis.bcaresearch.com. Cyclical Investment Stance Equity Sector Recommendations

Neutral When we downgraded the S&P home improvement retail index (HIR) to neutral in September, the core of our thesis was that fixed residential investment as a percentage of GDP appeared to have peaked, up 50% from trough to the recent highs, whereas relative HIR performance is up 170% in the same time frame (top and second panels).1 Such euphoria, both among investors and the sell side community (bottom panel), leaves the index prone to fall, should a disappointment occur. Yesterday's results from index giant Home Depot reinforce this point. The company exceeded earnings expectations and raised their guidance (albeit for only one quarter) and the stock still underperformed the market. Considering the cautious guidance from management on the housing market, cost pressures from tariffs and a much-reduced price of lumber, it is of little wonder the company failed to live up to expectations. Net, while we remain constructive on the overall housing market, we continue to think the positivity is fully baked in to the S&P HIR index; stay neutral. Clients seeking housing exposure should consider the compellingly valued S&P homebuilding index that we are overweight. The ticker symbols for the stocks in this index are: BLBG: S5HOMI - HD, LOW. 1 Please see BCA U.S. Equity Strategy Weekly Report, "Indurated," dated September 24, 2018, available at uses.bcaresearch.com.