The “improbable” election of Donald J. Trump as the 45th President of the United States has had a significant and positive impact on the financial markets. The market upswing has led many to wonder whether investors are getting carried away or whether the markets see something that analysts and economists are not seeing. For the latter, it is hard to explain the recent market behavior in terms of the economic fundamentals. After the election, economists surveyed by The Wall Street Journal added only 0.3 percentage points to their prior real GDP growth forecast for the next two years. At its December meeting, the Fed barely changed its real GDP forecast. In short, the establishment’s forecast would argue that if the fundamentals improved after the election, they did not do so by much. If not the fundamentals, what has changed?
We believe that the election of Donald Trump marks a policy inflection point. Even if tax rates and the regulatory burden do not decline by much, they denote quite a change in the trend of rising regulations and rising marginal tax rates produced by Obamanomics. If incentives matter, that alone is sufficient to produce an upward revision in the market’s valuation. The positive valuations also suggest that the market expects the tax rate reduction, regulatory relief, and infrastructure spending to have a positive impact on the pace of economic activity. One possible area of concern is foreign trade, where the mercantilist views expressed by the President are worrisome and we believe they will tend to offset the benefits of the other policies.
Identifying the Winners and Losers
At this point, we take the view that the Trump policies, overall, will have a positive effect on the economy. As time goes by, the details of the policies of Trump administration will be revealed and the policy uncertainty will be reduced. So far it seems that the president meant what he said and said what he meant. The uncertainty reduction removes one possible source of error in the design of a portfolio aimed at identifying the sectors and or stocks likely to benefit from the Trump administration. Where to start? During the presidential campaigns, it is customary for the brokerage houses and money managers to develop portfolios expected to benefit from the policies of the different candidates. That is a good starting point and, as suggested earlier, once the details of the president’s policies become better known, the portfolio is adjusted accordingly.
Traditionally, the selection of the stocks and asset classes identified as winners and losers focuses on the direct impact of the policies on the individual companies. Let’s take the case of a lower regulatory burden and a lower corporate tax rate.
Regulatory Burden: One simple approach to focus on the beneficiaries of the reduction in the regulatory burden is to identify the companies that are most affected by the regulations, as well as those least affected. The latter will be included in the loser’s groups, while the former will be designated winners of the regulatory easing. One can then broaden the analysis to focus on sectors with the most and least regulated companies. The ultimate extension—one that we favor—is to take a macroeconomic perspective and extend the analysis to an asset class. The lower regulations will favor U.S. companies relative to the rest of the world. Within the domestic companies, the smaller capitalization companies that cannot avoid the regulations will benefit the most from the lower regulations. Hence, early on our approach would favor the smaller capitalization stocks.
Corporate Tax Rates: A reduction in corporate tax rates will increase the after-tax cash flows of corporations and thereby increase their valuations, even if they do not change their behavior. The gains will be larger for the companies with the higher average taxes. Hence, that is the first step in identifying the stocks that will benefit from the lower tax rates. Also, to the extent that the taxation of foreign earnings is altered, information on the composition of the earnings may provide an additional screen. If one believes the “America First” slogan, then it seems reasonable to assume that the larger the share of domestic earnings, the greater the benefit from the America First policies. Putting it all together, the corporate screen is straightforward: focus on companies with the highest average tax rates and the largest share of domestically generated earnings.
Monetary and Credit Policy: In the aftermath of the financial crisis, the Fed adopted a policy of keeping interest rates low as far as the eye can see. We contend that in doing so, the Fed distorted the capital and credit markets. Now, as the Fed returns to normalcy during the Trump administration, the effects of these distortions will be reversed and those who benefited the most from the low interest rate policies are likely to underperform.
The Dollar and Protectionism: We contend that during the early stages the pro-growth policies will increase the expected real rate of return. That, in turn, will result in a capital inflow, a dollar appreciation, and an outperforming U.S. stock market. Unfortunately, the strong dollar will also fuel the protectionist fires and thereby increase the likelihood of protectionist policies. In our opinion, these policies will have a negative impact on the economy and the markets. The strong dollar and protectionist forces tend to cancel each other out. The impact on the economy will be the net of the two. Whichever one dominates depends on the portfolio of policies adopted by the Trump administration. If the economy picks up, we expect the dollar effect to dominate. In contrast, if the economy slows down, we expect that the protectionist forces have won the day.
Portfolio Construction Issues
In the previous section, we outlined several screens that we believe to be quite useful in the identification of stocks likely to benefit from the proposed Trump Administration economic policies. The screens are focused on the direct impact of the policy changes on the after-tax earnings (and thus valuation) of the different companies and asset classes. To obtain the individual companies’ estimates, one needs information on the earnings, their composition, i.e. domestic versus international, as well as the nature and amount of taxes paid.
The previous paragraph describes a good first step toward improving investment results. The downside is that there are data requirements. That data may not be readily available to some investors or they may not have the time to perform such an analysis. The question is whether there is a short cut? The answer is an affirmative one. We believe that a top-down, macroeconomic approach could provide a short cut that would get us to a superior portfolio without the need to collect all the individual company information required by the direct approach. It must also be pointed out that the direct approach does not take into consideration dynamic effects that may emanate from the policy changes being contemplated. The lower tax rates will result in an economic expansion and, all else the same, the firm that experiences the greater expansion will also experience the greater gain in profits.
Unfortunately, not everything else is the same. The ability to pass a price increase forward or backward will be critical to the ultimate impact on the profitability of the companies. For example, in the case of elastic sectors, where there are no barriers to entry, competition will dissipate the profits. In this case the consumers of the product will be the ultimate beneficiary. On the other hand, for inelastic industries that are shielded from competition through barriers of entry resulting from natural or technological reasons, they will retain the increased profits resulting from the lower taxes and regulations. These companies will experience an above-average performance and should be included in the Trump Portfolio. The question for the investor will be how to identify these companies.
There is an easy way to identify the companies that will potentially benefit from the Trump policies. One way is to let the market tell us. The answer can be found by looking at periods in the past where the economic environment has been like the one we expect. If a perfect match does not exist, we can look at time periods where the different indicators behaved the way we expect these indicators to behave as a result of the Trump policies. The rationale is that similar economic disturbances will have similar responses. Hence, those who outperformed during the relevant periods are likely candidates to outperform in the current environment. Next, we focus on the information we need to identify the relevant environment.
Our outlook calls for a rising slope for the yield curve and a declining spread for corporate bond yields. The expected increase in the slope of the yield curve reduces the likelihood that the long end of the yield curve will outperform the short end. The declining spread points to a relative outperformance of the corporate bonds over fixed income, suggesting that equities are likely to outperform fixed income instruments. Yet that does not answer the issue as to what happens to the inflation rate relative to the real GDP growth rate. Again, more information is needed.
The dollar reflects the price of two currencies and, under some general conditions, one can show that a dollar appreciation can be generated by one of three different variables: 1. A declining inflation rate relative to the rest of the world. 2. A rising real rate of return relative to the rest of the world. 3. A decrease in the risk premium relative to the rest of the world.
The dollar appreciation is a blow to the inflationistas, especially if one considers the fact that the rest of the world is not exhibiting rising inflation or rising inflation expectations. Given all the analysis in the press and recent international developments, it is hard to argue that the Trump victory brought calmness to the world economy and thus lowered the U.S. risk premium relative to the rest of the world. Thus, the process of elimination leaves the real rate of return as the logical explanation for the rising interest rates and a strong dollar.
The Stock Universe: We obtained total return information going back to 1993 for the current constituents of the Russell 1000. We then proceeded to eliminate the individual stocks that did not have returns for each year. That process left us with 284 stocks with a full set of annual data.
The Time Periods: The sample period for the different cycles is limited by our access to stock returns, which only goes back to the early 1990s. Yet during that time there is a wide range of experiences that lets us extract the information we need from mostly non-overlapping cycles. Looking at the data for the different economic environment indicators, we identified a mostly non-overlapping cycle for each of the indicators in question—the dollar, the slope of the yield curve, the corporate bond spread, and the Fed as far as the eye can see low interest rate policy. The time periods for the different cycles can be found in table 1.
The Classification and Performance
Once the cycles were identified, the performance of the individual stocks during each of the cycles was calculated and the stocks were ranked in ascending order and divided into 10 portfolios consisting of approximately an equal number of stocks. In cases where the cycle was a mirror image of our forecast regarding the environment we foresee during the Trump administration, we reversed the ranking.
The dollar cycle considered in the classification method was one of a declining dollar. It is the mirror image of what we expect early on during the Trump administration. Our analysis expects that those stocks that outperformed during the dollar decline cycle would underperform during the dollar increase cycle. Therefore, reversing the decile ranking obtained during the dollar decline cycle should be a good guide in the identification of individual stocks likely to outperform as a result of the dollar surge.
The average return generated by each of the deciles since November of 2016 suggests that the dollar screen does not reveal any systematic pattern. There is no significant difference between the top 5 deciles and the bottom 5 deciles. Does this mean that the dollar screen is ineffective? Or is there a possible explanation for the results? We lean towards the latter. If the Trump administration is concerned with competitiveness, whenever the dollar rises, the propensity of the administration to adopt protectionist measures will increase, thereby undoing the effect of a strong dollar on the economy and financial markets. The results reported in the first column of table 2 are consistent with this interpretation.
The slope of the yield curve cycle identified is one where the slope of the yield curve declined, the opposite of what we are looking for in the near future. Hence, as before, we expect that the stocks that outperformed during the declining slope cycle will underperform during the rising slope cycle. The performance of the decile ranking of the stocks in the universe are reported in the second column of table 2. The data is clearly consistent with our forecast. The half of the stock universe expected to underperform, i.e. deciles 1 through 5, posted a 9.52% average return versus a 10.77% average gain for the 284 stock universe. In contrast, the stocks in the decile expected to outperform gained an average of 12.02%, well above the sample average. More importantly, looking at the individual deciles, it is apparent that the first 4 deciles posted single digit returns, while the remaining decile posted double digit returns. The results suggest that there may be something to our approach.
The corporate bond spread cycle used in the classification is one where the spread rose, which is what we expect to see happening during the early part of the Trump administration. In this case a simple ranking of the decile will serve as a good proxy for the ranking of the individual stocks. The results of the screen are posted in the third column of table 2. The average performance is as expected. The first five deciles posted an average gain of 8.47%, while the portfolio consisting of the stocks in deciles 6 through 10 gained 13.09%. Comparing the spread results to the slope results, it is apparent that the spread portfolio generates a higher differential return, while the spread appears to produce more consistent results across deciles.
The return to normalcy cycle should be a mirror image of the experience of the last few years.
Again the data seems to support this view. The results reported in column 4, table 2 show that the portfolio consisting of the stocks in deciles 1 through 5 underperform the portfolio consisting of the stocks in deciles 6 thorough 10. However, the differential returns (9.61% versus 11.63%) are smaller than that produced by the corporate spread and the slope of the yield curve.
Putting It All Together
During the classification process, we took great care in making sure that the overlap among the different cycles was minimized. This way we hoped to capture the differential impact of the different environment or factors on the individual stock returns. We also assume that, to the extent that the cycles and factors are different, there is some degree of orthogonality. This is important because if the factors are different, then the different factors will characterize different components of the economic environment. By combining them, one may be able to replicate the environment we foresee.
One simple test of this hypothesis is to combine the different screens into a single screen and examine the performance of the portfolios. The results of this exercise are reported in the last column of table 2. The portfolio consisting of the stocks in deciles 1 through 5 underperform the portfolio consisting of the stocks in deciles 6 through 10. More importantly, notice that combining ranking of the different factors produces a portfolio of deciles 1 through 5 that generates a lower return than any of the other screens, while the decile 6 through 10 portfolio generates the second highest return of all screens.
This result supports the additivity and value added of combining the different screens.
One negative aspect of the results reported in the last column of table 2 is that the rank of decile performances is not quite what we anticipated. We expected decile 1 to post the lowest performance, decile 2 the second lowest, and so on. Is it possible that the relative ranking is greatly impacted by noise in the system? Recall that the dollar screen did not appear to be systematic. In order to test this possibility, we recalculated the rankings without the dollar screen. The results are reported in the next to last column. Notice that the relative ranking appears to conform to our expectations. While not perfect, the returns do rise as the deciles increase. This is reassuring. However, including the dollar screen produces a greater differential between the two portfolios in question. The moral of the story seems to be that for those interested in the broadest possible portfolio, the sum of all the screens is better.
Improving the Results
The methodology outlined here uses a top-down approach and we view it as a short cut to get us to a universe of stocks likely to benefit from the environment. The next issue is figuring out how to improve on these results. We believe that the answer is a bottom-up one. Knowledge of individual companies’ average tax rates, cost of capital, and share of earnings generated outside the U.S. can be used to significantly improve the results reported here.