Andrew Baron, Butterfield
Elections are, by their nature, uncertain, which is why they are held in the first place. If the U.S. was a homogenous society and a place of easy consensus, it would be quite straightforward to elect a president, even in a country of 300 million people.
It may appear that this election year has a larger degree of uncertainty and bears little resemblance to past elections, but it might do to remind readers that in 2008, Barack Obama was a rank outsider at the beginning of the nomination process. The country still wound up electing both its first African-American president, as well as one with a much more progressive agenda than had been popular for decades.
The one thing we can reliably count on is the likelihood that the pundits and pollsters will get it wrong! In that way, political punditry bears a striking similarity to financial market punditry.
So does the 2016 election matter to your portfolio? Is there a discernible link between the results (or predictions) of presidential elections and the economy? Further, is there a link between which party occupies the White House and financial market returns? In order to answer these questions – several academic studies and statistical surveys have tried – first there is the question of “belief.” Americans, perhaps all citizens living in democratic nations, have an inherent bias toward believing that their choice of political representative actually has some tangible effect on their lives.
Financial market professionals are, of course, not immune to this phenomenon. A 2012 election year survey by the CFA Institute of its U.S. members found that 80 percent of respondents believed that the outcome of the election would have an important effect, either positive or negative, on domestic economic performance.
The same survey asked whether it would have an effect on global economic performance, and 63 percent responded yes to that as well. There really is no way to tell who was right or wrong about economic performance, either by measurement of predictive ability, or in tangible data in hindsight. I can tell you, however, that if you ask a Republican whether sluggish global growth is Obama’s fault, the answer will likely be yes, and if you ask a Democrat whether a Romney presidency would have caused a recession, the answer would again surely be yes.
There are two basic, overriding problems with attempting to measure a presidential administration’s effect on economic growth. The first is simply that the executive branch, through the president, has little leverage over the economy. For example, both Federal Reserve policy and the price of oil exert much more influence over the course of growth and inflation than a president could hope to have. Fiscal policy is not even remotely in the president’s control, particularly in an environment like we have today, with one party in the White House and another controlling the Congress.
Secondly, most large changes in economic policy take many years, some perhaps a generation, to be effective and be evaluated objectively. Consider the following: Few would argue that the Obamacare legislation was a hallmark of President Obama’s administration and one that has the potential to affect U.S. economic performance over the long run. Over the course of the Obama administration, the economy’s trend growth has also fallen and productivity has decreased in what has been described by many as a “sluggish” recovery. It would be foolish (and statistically difficult) to conclude that Obamacare contributed to slower trend growth and lower productivity, but that will not stop many from trying to make a link of causality where one cannot be proven comprehensively. It is likely that decades of study will be necessary to understand the true effects of a policy change as large as a universal healthcare mandate for the labor force and many other policies may drive economic prosperity or cause headwinds in the intervening years.
As for the financial market effects, academic statistical research into whether Democratic or Republican administrations are better for stock market returns is, at best, inconclusive. From the 1980s until the early 2000s, conventional wisdom in academia and reams of market commentary was that better stock market returns were associated with Democratic administrations. However, a Federal Reserve paper in 2004 that adjusted returns for volatility found that the symbiotic link was tenuous. Some studies that measured from just before the Great Depression showed better returns for Democrats, but other research shows little difference in equity market returns by party going back to 1852.
As usual in statistics, the time period matters immensely, as does which data set one uses to come to a conclusion. The Leuthold Group and Ned Davis Research studied Dow Jones Industrial Average performance separated by each of the four years of a Presidential cycle, concluding that the period after mid-term elections, or a presidential administration’s third year, stock market returns rise most strongly. Still more research by Credit Suisse in 2011 prior to the last election found that S&P 500 returns since 1926 were much stronger when the incumbent party was re-elected, versus when the presidency switched to a new party “ownership.” In most all of the studies we have seen and the large amount of column inches expended on trying to prove one’s view, nearly every study is plagued by the fact that there have been many notable “exceptions to the rule” they are attempting to offer as a proof.
So what is the conclusion? Sadly, there sometimes is no conclusive answer and this is one of those instances. Evidence that “stock markets pick presidents” or that one party’s victory in a single election has a causal and immediate effect on the economy are sparse and riddled with potential for error and data-mining.
There is a saying in the investment business that accompanies any presentation of results, “Past performance is not necessarily an indicator of future performance.” Regardless of who wins in November, that statement will always hold true. History may repeat itself, but the reality is that sometimes it doesn’t, and you shouldn’t construct a long-term investment portfolio based on an election cycle or which television personality is sitting in a room without corners.
Sources: CFA Institute United States Presidential Pre-Election Poll – October 2012, CFA Institute, “Does the US Presidential Election Impact the Stock Market?” Lauren Foster 8/31/12, Electoral math: Presidential elections and the S&P 500, John McDermott, FT Alphaville 12/14/11, Stan Collender’s Capital Gains and Games Blog, Pete Davis, 11/4/08, Journal of Finance – Pedro Santa-Clara, Rossen I. Valkanov, 11/1/2003.