Should investors care about political campaigns? As we argued in the CIO View Special introducing our 2016 U.S. elections watch, it is easy to get carried away by rash campaign promises in the ups and downs of 24-hour news cycles. With one week to go, however, the time has come to take stock on what we have learned so far, as well as the potential implications for the economy as a whole, specific asset classes and sectors.
Heading into the home stretch of the campaign, Hillary Clinton seemed to have reestablished herself as the clear front-runner. Late-breaking developments, such as the latest revelations regarding Ms. Clinton’s ongoing email saga or Donald Trump’s tax returns, may not yet be fully reflected in the polls. These appear to be tightening in recent days. Still, Donald Trump is fast running out of time. Early voting is already underway in many key swing states. As of November 1st, 25m votes have already been cast, according to the University of Florida’s U.S. Elections Project.1
Source: RealClearPolitics: “General Election: Trump vs. Clinton”, Deutsche Asset Management Investment GmbH, as of 10/28/16
Mr. Trump’s uneven performance has also started to impact other races. According to most forecasters, Democrats now have a good chance to win back control of the Senate.2 However, it still seems likely that Republicans hold on to a majority in the House of Representatives.
If our base case materializes, we would expect markets to take a favorable view. In currency markets, we would expect some rebound on those currencies seen as most at risk from a Trump win, such as the Mexican peso. For bond markets, the main implication would be the removal of a potential source of uncertainty over monetary policy. Under a President Clinton, Janet Yellen stands a good chance of being reappointed Chair of the U.S. Federal Reserve Board (Fed), whose term ends in February 2018. (Any replacement appointed by Ms. Clinton would probably have broadly similar views.)
From a sector perspective, there is some risk of negative headlines on the health-care sector. However, most of Clinton’s proposals would need approval from Congress. Stock-price weakness out of fears, for example, of a Clinton administration penalizing drug companies may prove short-lived in this scenario.
In the longer term, a key question is whether a Clinton administration will be able to work more effectively with Congress than President Obama. A Republican House is likely to block many of the more eye-catching items on Hillary Clinton’s to-do list, including:
Trying to further expand Obamacare, notably by offering a public insurance option sponsored by the government;
Her various proposals to tax higher-income households, as well as to tweak corporate taxes;
More stringent financial regulation, including an annual fee on the liabilities of large banks and a tax on high-frequency trading.
All of the above should be understood as Ms. Clinton’s starting point for haggling with Congress. Ms. Clinton as well as Chuck Schumer, the incoming Senate Majority Leader if Democrats win, and Paul Ryan are all consummate deal makers. Speaker Ryan has already demonstrated as much, in his brief but surprisingly productive tenure so far. He will no doubt continue to be under pressure from his right-flank. Assuming he survives as Speaker with a reduced majority, allowing House Republicans some early wins may well be a sensible strategy for Democrats.
An obvious area to find common ground would be tackling the convoluted corporate tax code.3 The outlines of a potential deal have long been clear: limit deductions, close loopholes and cut tax rates. Any deal is also likely to include more favorable tax rules for the repatriation of foreign earnings currently stuck at overseas subsidiaries of U.S. companies. Beyond corporate taxes, we see some chance of a bipartisan compromise on infrastructure spending, which could provide a modest economic boost. (Ms. Clinton’s plans call for an extra $275bn in direct infrastructure spending over the next 5 years.) Progress may be harder in other areas, such as comprehensive immigration reform.
In addition to our base case, we have considered several risk scenarios. In choosing which scenarios to test, we relied on views from across the Deutsche AM investment platform, as well as outside data sources.4 For two key risk scenarios, we then used data collected bottom-up from across the investment platform, ran these through correlation-based models and evaluated the performance impact on various portfolios.
In the short term, a win by Mr. Trump could catch markets by surprise, create uncertainty and probably leave equity markets around the world under pressure. Credit spreads could widen. For U.S. Treasuries, we would expect a fall in yields, reflecting the murkier economic outlook. Many emerging markets look set to be hit especially hard, amidst fears of the U.S. turning protectionist. Currencies could experience sharp swings, with the Mexican peso likely to be especially hard hit. Against most major currencies, notably the euro, the U.S. dollar may also weaken sharply, while the Japanese yenshould benefit.
Beyond this immediate impact, it is difficult to speculate. Taken at face value, Mr. Trump’s plans would amount to a massive fiscal stimulus, at a time when the U.S. is already close to full employment.5 It would also likely lead to a sharp increase in U.S. debt issuance. But, as with most of his proposals, many key details to assess the precise impact are still missing. Getting any of his fiscal policies enacted would also depend on Mr. Trump’s willingness and ability to work with leaders in Congress. (Based on the trajectory of the race so far, a narrow Trump win in the Electoral College may well coincide with a loss in the popular vote.) Another source of uncertainty is who Mr. Trump would appoint for key positions, notably at the Fed.6
All this makes it hard to assess how lasting the negative impact of a Trump victory we expect to see in markets might prove. This presupposes a Trump administration prioritizing a more-or-less conventional Republican agenda of cutting taxes and regulations over rash campaign promises. (His proposals on foreign policy, trade and immigration would almost certainly hurt U.S. growth prospects in the longer term.) It is worth keeping in mind, however, that the two policy areas Mr. Trump has been most consistent on are opposition to free trade and immigration. There are also likely to be continuing headline risks in other policy areas. For example, Mr. Trump recently vowed “he’d break up the few corporations that dominate important industries”, according to Breitbart News, the right-wing website not so long ago headed by Mr. Trump’s campaign chief.7 Mr. Trump has repeatedly declared, notably in the first debate: “I think my strongest asset by far is my temperament.”8 It might take a while for markets to arrive at a similarly sanguine assessment.
What then, of the possibility of Democrats winning back both the House and the Senate on the coattails of a Clinton landslide have increased. Under this scenario, Ms. Clinton would stand a much better chance to enact many of the measures outlined above. We would expect this to be somewhat negative for U.S. equity markets, with health care, financials and energy likely to be the worst affected. These sectors might also see slightly wider credit spreads. For U.S. Treasuries, we could see a slight increase in yields. The immediate impact of all this on most bond and equity markets outside the U.S. should be fairly limited. The Mexican peso and the Canadian dollar could benefit.
However, market disappointment could prove relatively short-lived, not least as the majorities of Democrats in both the House and the Senate are likely to be very narrow. To have any chance of retaining them in 2018, Ms. Clinton may find it in her interest to reach out to moderate Republicans.
In addition to the above risk scenarios, we have also considered others. Notably, a tight race could result in legal challenges, creating uncertainty beyond Election Day. Combined with the recent strength of independent presidential candidate Evan McMullin in Utah, it could conceivably still result in all candidates falling short of the 270 votes needed to secure victory in the Electoral College.
At this stage, such scenarios appear too unlikely to test their investment implications. They highlight, however, just how unusual Mr. Trump has been as a candidate. It remains possible that polls are understating the support for him. Perhaps more plausibly, his complaints about election rigging may depress Republican turn-out and, his various controversies may encourage previously reluctant Clinton supporters to vote, notably among women, minorities and younger demographics. The remaining uncertainty cuts both ways, not least as it partly reflects a still unusually large number of voters who appear to remain undecided between the two unpopular front-runners.9 On top of that, Democrats have grown increasingly sophisticated in data analytics, building on their traditional strength in getting their vote out.10 At this stage, we would see the scenario of a Clinton landslide and a Democratic sweep not that much more implausible as that of a Trump win.
This also makes it hard to assess what exactly the market has priced in - which is one of the reasons investors should pay attention to political campaigns. So, how confident can we be in our assessment on the negative market implications of a Trump win? Once you think about it, this is a surprisingly tricky question.
One way to answer is to ask experts a few weeks before an event. This is basically what we did. One downside of this approach is that if and when an unexpected event materializes, it may well look and feel differently, not least to other market participants. Both market sentiment and the world at large will have changed in the meantime.
Another approach is to simply run a regression throughout the election campaign to see how financial markets react to trends in opinion polls or prediction markets. This is an excellent way to generate ideas, which is how we used it. Results need to be taken with a pinch of salt, however. There are usually plenty of confounding effects – macroeconomic data, for example, impacting both the asset prices and opinion polls.
To get around this issue, a third way is to look at specific, clearly delineated events during the campaign. Interestingly enough, a recent paper by Justin Wolfers and Eric Zitzewitz takes just such an approach. It looks at how well movements in financial markets during the first presidential debate reflected the odds of Trump winning according to prediction markets.11 According to their estimates a Trump victory would result in the S&P 500 Index being 12% lower than under Clinton, all else equal, and increase volatility. For stock-market valuations around the world, they see a similar impact of 10 to 15%. In our view, their figures probably overstate the likely impact, and of course, one should not read too much into a single event.12 Still it is reassuring from a modelling, if not from an investment perspective, that all three approaches yield similar conclusions for a wide range of assets.
1 For regular updates, see http://www.electproject.org/early_2016
2 See for example, http://www.nytimes.com/interactive/2016/upshot/senate-election-forecast.html, which also contains a useful comparison of the results from various forecasting models.
3 This is an area “Ripe for Bipartisan Reform”, as Arthur Laffer, a former advisor to President Reagan, put it in a recent research report with the same title. https://www.pacificresearch.org/fileadmin/documents/Studies/PDFs/2013-2015/NewCorporateTaxF.pdf
4 To regularly update our thinking on the race, we pay particular attention to the 538 Polls Plus forecast by Fivethirtyeight (http://projects.fivethirtyeight.com/2016-election-forecast/?ex_cid=rrpromo#plus) and data from prediction and betting markets via PredictWise (http://predictwise.com/politics/2016-congress-house).
5 Mr. Trump has called for slashing both corporate and personal taxes, and ruled out tackling entitlement spending. He presumably wants to increase defense spending, but only after conducting a “full audit” of the Pentagon. When asked about Ms. Clinton’s plans on infrastructure, he stated in August: “I would say at least double her numbers, and you’re going to really need a lot more than that.”, according to Berman, Russell, “Donald Trump's Big-Spending Infrastructure Dream”, The Atlantic, 8/9/16 http://www.theatlantic.com/politics/archive/2016/08/donald-trumps-big-spending-infrastructure-dream/494993/
6 Part of the problem on this and many other economic issues is that Mr. Trump’s positions have been inconsistent throughout the long campaign. On the one hand, he has repeatedly rebuked Ms. Yellen for keeping interest rates low. On the other hand, he has long appeared to favor low interest rates, for example declaring on May 5th: ““I am a low interest-rate person. If we raise interest rates and if the dollar starts getting too strong, we’re going to have some very major problems.”, according to Davidson, Kate, „Donald Trump’s Shifting Words on Interest Rates and Janet Yellen”, The Wall Street Journal, 9/12/16, http://blogs.wsj.com/economics/2016/09/12/donald-trumps-shifting-words-on-interest-rates-and-janet-yellen/
7 McHugh, Katie, "Trustbusting: Trump vows to break up the new media conglomerate oligopolies", Breitbart News, 10/24/16 http://www.breitbart.com/2016-presidential-race/2016/10/24/trustbusting-trump-vows-to-break-up-the-new-media-conglomerate-oligopolies/
9 For a brief but excellent overview of the sources of uncertainty and various modelling approaches, see Silver, Nate, “Election Update: Why Our Model Is More Bullish Than Others On Trump”, Fivethirtyeight, 10/24/16, http://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/
10 Goldmacher, Shane, “Hillary Clinton’s ‘Invisible Guiding Hand’”, Politico Magazine, 9/7/16, http://www.politico.com/magazine/story/2016/09/hillary-clinton-data-campaign-elan-kriegel-214215
11 Wolfers, Justin and Zitzewitz, Eric, “What do financial markets think of the 2016 election?” The Brookings Institutions, draft as of 10/26/16, https://www.brookings.edu/wp-content/uploads/2016/10/what-do-financial-markets-think-of-the-2016-election_102016_wolferszitzewitz.pdf
12 This is reflected in the range of the Wolfers and Zitzewitz estimates: 6 to 18% for the S&P 500 Index, at a 95% confidence interval. Our own bottom-up assessment of the market reaction in the wake of a Trump victory would be slightly below the lower end of that range for the S&P 500 Index. Partly, this is because the relationship between the probability of a Trump victory and stock markets’ reaction is probably not quite as simple as the paper assumes. (For example, it does not capture the growing risk we see of the negative impact a Democratic sweep might have on market sentiment.)