The economist and Nobel laureate Friedrich August von Hayek famously described markets as a “marvel.”1 Taking in information distributed among millions of individuals from all around the world, markets are able to aggregate it into price signals to guide decision-making. When it comes to the 2016 U.S. election cycle, those market signals are pretty hard to read, however.
Anecdotally, investors appear increasingly concerned about political risks ahead. This seems understandable enough. After all, the Democratic nominee and front-runner, Hillary Clinton, continues to be dogged by her ongoing email saga. She has barely given a news conference since the whole tedious process of getting elected got started. Meanwhile, her opponent Donald Trump has caused concerns, especially overseas, with his skepticism of free trade. On other topics, he continues to surprise friend and foe alike, including his own surrogates, with changing positions.
All this sounds like a recipe for uncertainty of the sort that might cause market jitters. Instead the S&P 500 Index is trading near all-time highs!
This seems odd, not least as history suggests that U.S. equity markets are generally slightly weaker in election years, with share-price gains averaging 6.5% since 1936, compared to 8.7% in all years since 1933. In election years since 1936 when there is an open presidential contest, with no incumbent on the ballot, the S&P 500 Index has generally been weak, showing, on average, slight price declines.
Of course, such statistical results should come with a soup spoon of salt. There are simply not enough recent election years to make any reliable inferences. The relationship between the economy and financial markets is complex and ever-changing. Worse still, there are breaks in the data when it comes to how electoral politics might impact both the economy and markets. In the statistical jargon, there are plenty of non-stationary processes involved. (To take just one example, consider how U.S. monetary policy changed from 1979 on, when the Federal Reserve (Fed) under Paul Volcker got tough on inflation.)
One powerful way to make that point is to reconsider the above chart – and, instead of taking the average year-to-date performance in open election years since 1936, exclude 2008. That alone lifts price gains from -3.3% to 3.7%. Now, consider excluding 2000 as well. For the remaining other open election years since 1933, performance was pretty much in line with that of all election years.
One way to read the current mood in markets is to imagine that investors are already pricing in a continuation of the status quo. This would mean Democrats controlling the White House, with Republicans remaining sufficiently strong to act as an effective counterweight in Congress.
For now, such an outcome looks plausible. At the time of writing, the Democratic nominee Hillary Clinton has emerged as the frontrunner (see chart below). Simple averages of recent polls show Mr. Trump lagging. Ms. Clinton secured a rather more sizeable and sustained boost from her party’s convention in late July than the real-estate mogul had received a week earlier.
Based on more sophisticated analyses, most observers currently only give Mr. Trump a slim chance of winning – in the region of 33% or lower, depending on which model one looks at.2 As described below, however, such predictions mask plenty of uncertainty.
Polls suggest that voters find both leading candidates rather uninspiring.3 After the recent Brexit surprise, moreover, we have found a renewed interest in tail risks among investors. In this Special, we argue that it is too early to get nervous. However, investors should keep an eye on the risks ahead. The race could yet prove more volatile than markets appear to be expecting. Moreover, a key question will be who controls the House of Representatives and the Senate. Arguably, this should matter as much, if not more, to investors than the headline-grabbing presidential contest.
Labor Day, the first Monday of September, traditionally marks the time when ordinary voters begin to zoom in on the presidential race. It is also the stage when analysts, fund managers and strategists increasingly start to fret about the impact on various assets and sectors.
Of course, old hands on Wall Street know that it is all too easy to get carried away amid the whirlwind of proposals, debates and sound bites. The U.S. Presidency may still be the most powerful job in the world; except for foreign policy, however, there is little a President can do without congressional support. When it comes to the things investors care about most, Congress has the power of the purse – an important reality of which the markets were reminded several times during the Obama years amidst budget standoffs and looming government shutdowns.
America’s famed checks and balances are not the only reason why it is unwise to make long-term investment decisions on the basis of rash campaign promises. As the Brookings Institution, a Washington think tank, once noted in a primer on presidential transitions: “Every new President wants to hit the ground running, but none wants to run alone.”4 The sheer logistics of getting a new administration in place are mindboggling.
The challenge, this time around, might appear particularly overwhelming for Donald Trump, not just the de facto leader of a party out of power for eight long years, but also a political novice. Unlike previous newly elected Presidents in a similar predicament, he lacks a deep network of contacts among policy wonks in Washington and beyond. It is not just that one does not know what he might really think on the sort of specialized topics that might interest an investor from the perspective of a particular sector. It is not even clear whom he or his people are likely to ask. Mr. Trump has been on a warpath against many of the policies pursued by the most recent, previous Republican administrations. This probably rules out bringing back the old team, a favorite fallback option to smooth presidential transitions. Identifying enough individuals in sync with Mr. Trump’s priorities to fill thousands of mid-level positions may prove to be complicated.
A rough rule of thumb is that it takes the better part of a year until a new President’s team is firmly in place – and often longer for key, but controversial appointments. By the time the new administration is working properly, the world has usually moved on. Priorities change and events have a tendency to derail even the most clearly stated ambitions.
(For example, first-time presidential candidate Obama promised to renegotiate the North American Free Trade Agreement (NAFTA)and to crack down on Chinese exports, accusing China of manipulating its currency.5 After the election, he largely embraced a pro-trade agenda. And remember how George W. Bush ran as a compassionate conservative in 2000 and promised to steer clear on foreign entanglements?)
All of this is why we think it would be premature to draw detailed implications from the race so far for the economy as a whole, specific assets or sectors. (For example, Mr. Trump has yet to clarify his plans on taxes, having recently withdrawn his previous draft proposal from his campaign website.)
We will do so in coming weeks, as both candidates’ likely priorities and chances of securing a convincing mandate become clearer. By that time, it may also be possible to assess the chances of control in the Senate and – more remotely – the House of Representatives changing hands from Republicans to Democrats. The latter is of particular significance for investors, as only members of the House of Representatives can introduce bills concerning taxes.
In the meantime, the rest of this Special explores the state of U.S. politics and the longer-term implications of the turmoil we have recently experienced. Specifically, we will look at:
I. Democracy in America in 2016 II. Governing a changing nation III. The state of the race and of electoral forecasting IV. Market implications in the weeks and months ahead
For several election cycles, U.S. voters have been understandably upset about Washington’s chronic inability to get much done. In recent years, such complaints have only grown louder.
It is not hard to see why there is so much dissatisfaction. By many measures, Washington has never been as dysfunctional. A deeply divided government may arguably be a proxy cause of this. A Democrat has held the White House for 16 of the past 24 years. By contrast, the House of Representatives has mostly been in Republican hands since 1994, when Newt Gingrich ended almost 40 years of Democrats controlling the lower chamber of Congress. That only begs further questions however. Not so long ago, after all, the roles were reversed. Between 1968 and 1992, Republicans typically had an edge in winning the White House, but were weak on Capitol Hill. This did not prevent Congress from granting Richard Nixon the power to impose wage and price freezes in 1970, nor did it derail Ronald Reagan’s economic agenda in the 1980s.
America’s founding fathers were assiduous in avoiding any one person or branch gaining too much power – hence, the emphasis on checks and balances. The idea behind having two chambers was that representation in the House should be based on population. By contrast, six-year, overlapping terms in the Senate would act as a brake on rash schemes. Having two Senators each would also ensure that small, less populous states could make their voices heard in Washington. This is one reason for what may appear to an outsider as oddities of the U.S. political system.
In the U.S. electoral system, oddities abound. The world was reminded of that in the 2000 election, when Al Gore won the popular vote, but ultimately lost the Presidency after the Supreme Court stopped a recount in Florida. Most states award electors on a “winner-takes-all” basis through the Electoral College, which chooses the President and the Vice President. (The exceptions are Maine and Nebraska, where some electors are picked at the level of congressional districts.)
Casual observers might be less familiar with some other quirks of the U.S. political system. In particular, the number of electors in the Electoral College is based on the total number of Senators and Representatives. Since each state – no matter how small its population – has at least one Representative and two Senators, each gets at least three electors. This gives somewhat disproportionate influence to small, mostly rural states.
Currently, seven out of 50 states have only one Representative (as well as the standard two Senators): Alaska, Delaware, Montana, North Dakota, South Dakota, Vermont and Wyoming. And, as below chart shows, there are plenty of other small or sparsely populated states with fewer than one million residents per U.S. Senator.
Of course, the impact of this federal structure is much larger when it comes to the legislative process than it is in the Electoral College. Each Senator from a state like (tiny) Vermont, say, has just as much voice as the Senators representing the most populous states, such as California, Texas or Florida. In particular, he or she can dramatically slow down proceedings by withholding unanimous consent. This contributes to the impression of gridlock in Washington.
Incidentally, residents of Washington, D.C. have rather less of a say than you might think. Because it is not a state, America’s capital has no voice in the Senate and does not even get to run its own affairs. It does get to send non-voting delegates to the House and has three electors in the Electoral College.
Democrats appear to have a decent chance of winning back the Senate. However, whoever wins the presidency will probably have a Republican House to deal with. In our view, it is too early to speculate how this might work in practice. On the one hand, U.S. society is indeed changing, as we will see in the next section.
On the other hand, personalities and leadership matter. This includes President Obama, who has been unusually detached from the legislative process. As for Congress, it is hard to dispute the evidence that under Republican control, the House has tended to struggle to properly function as a legislature.6 Some, such as political scientist Thomas Schaller, go further. In his book The Stronghold, he argues that “It is Gingrich, not the lionized Ronald Reagan, who should be remembered as the most significant Republican politician of the late twentieth century.” 7 According to such views, the changes put in motion by Newt Gingrich and others since the 1990s have permanently changed the incentives of Republicans. Even if this is correct as a historical observation, it is worth keeping in mind that Gingrich resigned as speaker in 1998, almost 20 years ago. A lot has changed since then.
Governing a changing nation
Governing a changing nation
In demographic terms, the U.S. society is changing. Older Americans are still overwhelmingly white (or Caucasian, in Census terms). Among children under five, other ethnic and racial groups are already the new majority.8
No doubt foreign readers will hear plenty about this ahead of November 8th. Minorities tend to overwhelmingly vote for Democrats. Donald Trump may have made matters worse for the Republican Party with numerous comments that appear to have caused offence.
In our view, the importance of demographic change should not be overstated. In fact, there are plenty of reasons to be cautious on the alleged size of a demographic boost Hillary Clinton might receive, as we will see when we examine the state of the race.
Demographic change has, however, already had one particularly significant consequence. Minority and younger voters generally tend not to vote during mid-term elections – when all of the House and one-third of the Senate are up for grabs. Turnout is much higher when the presidency is also on the ballot. As result, the electorate is very different – younger and more diverse in presidential years, compared to a whiter and older electorate in mid-term elections. This is one of the reasons why Congress has different electoral incentives than the President.
Another is that increasingly, Americans live alongside people similar to themselves. In his 2008 book, Bill Bishop called this The Big Sort, with the telling subtitle: Why the Clustering of Like-Minded America is Tearing Us Apart.9 The provocative thesis in a nutshell is that Americans are self-segregating in terms of how they eat, think and worship. And, that appears to extend to how Americans vote.
One result is increasingly partisan neighborhoods – and a growing lack of understanding of how the other half of the electorate think and live.
This has also worsened a long-standing problem of American democracy. Every 10 years, there is a redistricting of congressional districts in the House of Representatives to reflect population changes, the last of which was based on the 2010 Census results. In most states, it is a thoroughly partisan affair. Since the 1980s, better and cheaper mapping software has also made it increasingly simple to manipulate congressional district maps.
Whenever they are in control, both parties have generally tried to ruthlessly gerrymander districts to maximize their overall electoral advantage. The results are districts whose geographic shape defies logic – and more and more Representatives elected in increasingly partisan districts. Such parliamentarians have little to fear from the general election – but everything from primary challenges from within their own party. This, in turn, has changed the incentives of both parties on Capitol Hill and the way Congress works.
This currently works in favor of Republicans, who control most state legislatures in charge of redistricting. For example, in the 2012 and 2014 elections, Republicans received about 3.8 and 4.4% more House seats, respectively, than their national margins suggested.10 Gerrymandering can explain only part of this – another factor is how concentrated Democratic votes are in urban areas. In this regard, this year’s elections for the House of Representatives will be instructive. In several key states, notably Florida, congressional-district maps were ruled unconstitutional by state courts. As a result, a far wider range of congressional districts is likely to be competitive this time around.11
Nevertheless, chances of Democrats winning back the House look remote for now. In part, this is because a big lead for Ms. Clinton in the polls might simply encourage ballot splitting. Some voters might choose her for the Presidency – but deliberately vote for Republicans in Congress to act as a counterweight.
Before betting on such an outcome, however, investors would be well advised to take a closer look at the state of the race - and that of electoral forecasting in the U.S.
The state of the race...
The state of the race...
Remember the days when weather forecasters were the butt of jokes? If you are in your thirties, or older, you probably do. Over the past 40 years, weather forecasting has progressively gotten better.12 The reasons why are described in one of the chapters of Nate Silver’s fascinating and entertaining book, The Signal and the Noise: Why So Many Predictions Fail – but Some Don't, which takes a look at all sorts of things people try to predict, from hurricanes to stock markets.13
Something similar to the progress seen with the weather has been happening in the field of U.S. political forecasting. It may sound hard to believe in the year of such notable forecasting failures as Mr. Trump winning the Republican nomination and the United Kingdom voting to leave the European Union. As a matter of fact, however, sources for forecasting elections have gotten much better over the past decade. Nate Silver was one of the pioneers of data-driven analysis when it comes to elections (and sporting events before that, where quants have transformed not just sports betting but the sports themselves14 ).
One of Silver’s starting points was the ground-breaking work by political psychologist Philipp Tetlock. Based on a series of forecasting tournaments between 1984 and 2003, Tetlock tried to figure out two things. First, how good was the average political expert at making forecasts? And second, were some experts better forecasters than others?
The answers to both questions were quite surprising. It turned out that a vast number of well-paid experts were no better than an algorithm randomly assigning probabilities. Provocatively but perhaps unwisely, Tetlock described this finding as the average expert being no better at forecasting than a dart-throwing chimp. That rather detracted from his second, arguably more significant finding. Some experts were actually quite good, and they appeared to share certain cognitive characteristics.15
Tetlock called those good forecasters foxes, borrowing the term from Isaiah Berlin’s funny little 1950 essay The Hedgehog and the Fox, who in turn was inspired by the fragment of the ancient Greek soldier-poet Archilochus: “The fox knows many things, but the hedgehog knows one big thing.”16 A fox is someone like Nate Silver, who draws from a wide range of sources, has few preconceptions and is disciplined in keeping track of past forecasts, so that accuracy goes up over time. In other words, willing to change his mind as frequently as the evidence seems to suggest, but not too frequently.
Tetlock suspected that such mental habits should prove useful across a wide range of forecasting domains. Silver’s book suggests that this is indeed the case. Tetlock also believed that such traits could be learned and that combining solid forecasters into teams might boost performance even more. He sums up his results in the recently published and highly accessible Superforecasting: The Art and Science of Prediction17 (which should arguably be recommended reading for anyone making predictions for a living, including those who manage other people’s money).
But if all this is so, why did so many people get it so wrong in the case of Mr. Trump, or indeed Brexit. In May, Silver gave his answer in a posting on FiveThirtyEight, the website he founded and runs: “How I Acted Like A Pundit And Screwed Up On Donald Trump.”18 Most pundits have the opposite set of cognitive traits than those that foxes display.
In Tetlock’s terms, a typical pundit is a hedgehog – someone who has one or a few big ideas, conceptual models or core beliefs and applies them to any forecasting problem that comes along. If those forecasts fail to materialize, hedgehogs tend to double down on their analysis, even in the face of mounting evidence that they might be wrong. Worse still, they might even avoid undertaking the sort of tests that risk proving a pet forecast wrong.
As Silver explains, that is more or less why he underestimated Mr. Trump for far too long. The polls were, in fact, suggesting that Mr. Trump was on to something. It’s just that because a Trump win seemed so unlikely, Silver was reluctant to trust what the poll numbers were telling him. As the below box describes, he had good reasons to do so.
Remember how pundits were sure that Donald Trump was never going to amount to much once the voting actually started? There were good reasons to believe the conventional wisdom. These are summed up in the influential 2008 book, The Party Decides by political scientists John Zaller, Hans Noel, David Karol, and Marty Cohen. Their argument is that for all the fanfare that surrounds primary campaigns, the stakes are, in fact, stacked against outsiders.19
In addition to the ability of charming and inspiring ordinary voters, any potential candidate needs a deep war chest or enthusiastic support among the grassroots or influential allies in partisan media and preferably all three. So, before presidential candidates can even begin to make their case to Iowa voters (traditionally the first state to vote), they have to pass through an invisible primary. This involves reaching out to TV pundits and policy wonks, gaining endorsements from fellow politicians and raising money. As a result, the winner is usually one of the candidates thought to be able to succeed in the general election, while at the same time being more or less in sync with party opinion. Party elites, after all, are interested in winning and retaining power. And when forced to choose, they tend to support seemingly electable candidates such as Mitt Romney or John McCain over fire-brand rivals.
In 2016, this argument has been much discussed, especially as Mr. Trump started gaining ground. Showmanship and deliberate bluntness allowed him to dominate the airwaves, depriving everyone else in a field of 17 (serious) candidates of the oxygen of publicity. More importantly, he zoomed in on certain key elements he rightly thought would appeal to Republican base voters, which had been neglected by his rivals and other office holders. Opposing free trade gave him a unique selling proposition. That put other candidates in a difficult spot when trying to make a credible counter offer to voters without losing business support.
On everything else, it appears that Mr. Trump has been willing to move – and take whatever position seemed to trigger the strongest emotive reaction from his audiences. Democrats might underestimate him at their own peril in the general election.
After all, factors not captured in the polls – notably the lack of party endorsements that underlined Mr. Trump’s weakness in the invisible primary – seemed to suggest Mr. Trump would underperform once the actual voting started. And at the time this was not wholly unreasonable. The invisible primary had been key in understanding many previous campaign cycles.
Something similar happened with Brexit. Heading into the referendum the polls were actually reasonably accurate – according to an average of polls published in the Financial Times, they were showing a 50/50 split on the eve of the vote. It was just that many people refused to trust the numbers – and had some good reasons to do so. Notably, betting markets suggested a clear win for “Remain”. Such markets had lately tended to be more accurate at predicting UK voting behavior than polls (which, for a variety of reasons, have lately had a patchy track record in the UK).
So, where does that leave us, as we turn to the current state of the U.S. elections?
Well, first of all, it suggests that we should take seriously what models with a solid track record tell us, but not trust them blindly. In the case of U.S. general elections, we would argue that past performance does justify having some faith in the Fivethirtyeight forecasts, in particular their polls-plus forecasts, which uses a weighted set of polls as well as demographic, economic and other factors that have proved reliable in the past. This currently suggests a 33% probability of Donald Trump winning the presidency.20 Fivethirtyeight also publishes polls-only forecasts for the U.S. as a whole, as well as individual states. This relies solely on polls weighted by reliability, sample size and recency and puts the probability of Trump winning at 30%.21
Other models, such as that of the Upshot in the New York Times, suggest an even more difficult path for Mr. Trump. The Upshot takes a similar approach to the Fivethirtyeight polls-only forecast, but uses different poll-weighting methodologies and a somewhat more conservative approach in incorporating trends from new polls. It currently puts the probability of Mr. Trump winning at just 18%.22 The main reason for this significantly lower probability appears to be that the Upshot tends to assume the results from key swing states will be less correlated than in the Fivethirtyeight model. This makes a Trump victory through a series of upset victories look less likely.
However, all such figures need to be treated with caution. This is because Donald Trump’s coalition is very different from that of past Republican candidates – increasing the scope for errors in both polling data (sampling biases, over-/under-reporting, etc.) and applying national data to a map of potential swing states.
In addition, polls are consistently showing relatively strong support for Libertarian party candidate and former governor Republican Gary Johnson. This could yet result in him being included in the televised debates, which would require consistently winning 15% in national polls. According to recent averages of national polls, he is currently at around 9%, while Jill Stein, the Green party candidate, is at approximately 3%.23 Given historically low favorability ratings of both Mr. Trump and Ms. Clinton24 , it is difficult to predict who will be hurt more by Mr. Johnson, Ms. Stein and other alternative candidates in key swing states.
Another difficult set of questions is whether Ms. Clinton will be as effective as Barack Obama in turning out her base. During the long primary contest with Bernie Sanders, her main opponent for the Democratic nomination, she struggled to inspire support among younger voters, a key part of Mr. Obama’s coalition. It is difficult to assess, moreover, whether turnout among minority voters will be quite as high as when they were electing and re-electing the first African-American president. In part, this might depend on whether Donald Trump pivots towards a more inclusive campaign style, which might also help reduce Ms. Clinton’s lead among female voters.
In brief, there remain plenty of uncertainties, justifying a wide range of plausible forecasts. Most notably, plenty of voters remain undecided. This number should shrink as the election nears, making it somewhat easier to make reliable predictions.
So, where does that leave us for the rest of the race? And what might happen in the markets in the weeks and months until Election Day and beyond?
The short answer is that it is too early to tell. Based on recent dynamics, we can see two reasonably plausible scenarios for the coming weeks:
If this were to happen, it would suggest that Democrats have succeeded in making Mr. Trump look unqualified to be Commander-in-Chief. Based on past experience, the debates are unlikely to provide the sort of blow-out performances that might change that and would probably reinforce existing dynamics. Ms. Clinton has had plenty of experience in debating a single opponent. However, the very fact that Trump is poised to enter the stage as the underdog could work to his advantage.
If Ms. Clinton retains her lead after the debates, markets may well treat the rest of the campaign as a non-event. Moreover, certain sectors expected to suffer under another President Clinton, such as health care or oil and gas, may continue to underperform, while others, such as green energy, might benefit.
The overall equity market may be more influenced by global events and Fed speculation than anything happening in the political news. However, those expecting smooth sailing under another President Clinton might want to dust off a copy of Game Change, the 2010 book by John Heilemann and Mark Halperin and probably the definitive account of the 2008 race – as well as the turmoil inside the Clinton campaign.25
In this case, expect plenty of chatter on how the U.S. market might catch a political fever. An additional cause for concern might be fears of an October surprise, such as new, incriminating emails being released by WikiLeaks, just ahead of the election.
However, it is far from clear that a close race would permanently weigh on market sentiment. For one thing, if Mr. Trump feels he is on track to win, this could perhaps make him less erratic, less threatening and more inclined to moderate his positions. It is also quite possible that market participants will start to anticipate potential positives from fiscal stimulus or infrastructure spending, for example, under a Trump Presidency.
If we head into Election Day with the race still very close to call, there are good reasons to think the forecasts (even good ones) might get it wrong. Due to Mr. Trump’s unusual positions for a Republican candidate, some states might yet flip unexpectedly. Moreover, close races in unexpected places, combined with recently enacted voting restrictions (such as voter-ID laws) could cause delays in the count or even prompt court challenges. A close election might well keep markets on edge until well after November 8th. That said, keep in mind that Democrats have invested far more heavily in getting out the vote.
And beyond November 8th? In particular, what would happen if Mr. Trump wins? The short answer is that at this stage it is hard to tell what this would mean for the economy and markets. The main issue is that it remains impossible to know what Mr. Trump really wants. His impulsiveness and unpredictability have arguably been key factors underpinning his success. For one thing, they have made Trump a favorite on social media well beyond U.S. shores. Starting last year, the Republican presidential debates became an overnight internet sensation in China. They generated millions of views on Weibo, China’s leading microblogging site, after a few Chinese student volunteers had the bright idea to subtitle the first one, on August 6th, 2015. “Watching the U.S. presidential debates is like watching a football match,” one of the volunteers observed. Other users praised the “very sophisticated and wonderful fight.”26 Given the stakes involved, however, Mr. Trump will swiftly have to prove to the U.S. electorate that he can act as presidential as he has long promised his supporters.
Things might get clearer as Election Day gets closer, and we get a better feel for House and Senate composition. For now, we leave you with three observations.
First, the Trump experience may permanently change the Republican Party. All along, Republican thinking has been dominated by pro-market enthusiasts such as Hayek. The era of such ideas holding sway in Republican circles may be approaching its twilight. Their influence will not vanish overnight, though. Trump’s success aside, this year’s primary election results for down-ballot races do not suggest a broad-based movement for more populist policies. Even if Mr. Trump loses, however, others may try to use his playbook. Already, Mr. Trump’s candidacy has significantly eroded Republican support for free trade, once a core article of faith among Republican policy-makers.
Second, at least until the dust settles, investors should be cautious in applying old rules of thumb. For the past 36 years, Republicans have reliably tended to pursue market-friendly supply policies, while showing themselves pragmatic on countercyclical spending in times of economic duress. It is less clear that this will continue to be the case. A Trump victory might, for example, increase existing congressional pressures to tighten the Fed’s scope for maneuver. Across the board, policy-making would probably become harder to predict. Similarly, the Bernie Sanders insurgency has dragged Hillary Clinton to the left. It remains to be seen whether she successfully pivots to secure a mandate for business-friendly policies along the lines her husband pursued.
Third, keep in mind that the general election campaign has barely begun. Weird things do happen during campaigns, as CBS moderator John Dickerson reminds us with each one of his Whistlestop podcasts, recently released in book form.27 For now, caution is advisable in assessing which campaign promises (or threats) investors can take seriously – and which to disregard.