If politicians frequently ignore the impact of their policies on the economies of the countries they govern, (we will discuss why in future columns), it is also fair to say that economists have mostly ignored the fundamental role of governments in determining economic outcomes. For much of the 1950s to the 1980s, growth economists didn’t consider governments at all when working on basic questions such as ‘Why are some countries rich and some countries poor?’
In the past decade, this sorry state of affairs has changed and political explanations of variation in levels of prosperity have become mainstream. While very convincing at one level, it is going to be very hard to establish whether these explanations are correct. The difficulty lies in establishing causality, how one thing actually leads to another, a problem that bedevils the social sciences.
The so-called Presidential Puzzle (first documented in an academic paper in 2003 by Santa-Clara and Valkanov) is an excellent example of the problem. You may not be aware that, if the future resembles the past, you should invest in US stocks only when the President of the US is a Democrat. When the President is a Republican you would be well advised to leave your money in the bank and go fishing.
Between 1926 and 2016, the average annual return on a comprehensive index of American stocks in excess of the one-month Treasury Bill was an impressive 10.7% during Democrat presidencies. At such rates of return a one-dollar investment doubles in value roughly every seven years, and increases over one thousandfold after seventy. Under Republican Presidents the stock market average excess return was a pitiful 0.2%. Economic growth was also higher, with GDP growing on average at 4.7% per year under Democrat presidents since Theodore Roosevelt and 2.7% under Republicans since then.
These differences cannot be explained by random statistical chance, hold in the sub-period since the end of the Second World War (so are not explained by the Great Depression and the war), and have grown stronger since the publication of Santa-Clara and Valkanov’s original study nearly 20 years ago. The two parties occupied the White House for roughly equal amounts of time over all of the periods in question and alternated regularly.
Motivated by this and other evidence, economics researchers have speculated that there may be partisan effects on the economy, paradoxically that for reasons that are not fully understood the supposedly pro-business Republicans are bad for economic growth and stock market values whereas the business-antagonistic Democrats are the opposite. But before we start speculating on what obscure and surprising forces are at work, there may be a more fruitful line of enquiry.
The US electoral and party system is such that neither party enjoys an entrenched majority. Both sides have their base, their core vote which remains loyal, and each party seeks to attract a relatively few swing voters who tend to be more centrist than the base of either party. Even quite moderate swings can result in very large margins of victory for one side relative to the norm, as happened in the elections prior to Reagan’s and Obama’s second terms.
What drives these swings? Well, many different things, but one obvious factor would seem to be economic conditions. The economic times produce voters of a particular outlook: The Great Depression caused one third of American men to be unemployed, times were very hard, and this created millions of swing votes in favour of progressive policies. The result was that Franklin Roosevelt won the election. After the Great Depression’s terrible beginning, the US economy grew again and the stock market recovered.
On a more everyday level, Bill Clinton won his first term with the catchphrase ‘It’s the economy stupid’ when the US was technically in a recession, but the mildest recession it has ever experienced. The economy and stock market went on to the internet boom of the 1990s.
In other words, it is much more likely that the causality works in reverse: bad times change swing voters from Republican to Democrat, and vice versa, and bad times, at least in the United States, tend to be followed by better times, and quickly too. When the median voter is fearful, she votes Democrat for that welfare check, stocks are cheap, and investment is low. When she is bullish, she votes Republican for lower taxes, stocks are expensive and investment booms. The high returns are made during the transition form bear to bull.
So does it really matter, for the economy, whether a Red or a Blue candidate sits in the White House? Or is the Presidential Puzzle really just an outcome that reflects the fact that the mood of swing voters determines both election outcomes and subsequent economic outcomes in the US when their mood changes, as it invariably does?
It’s very difficult to tell, but I think that the policies of Presidents do appear to affect the economy. The reason is that when the same party controls both the White House and both houses of Congress, we see a similar phenomenon to the Presidential Puzzles, but regardless of which party controls the White House. Contrary to received wisdom, since 1926 stock market returns were on average 10% per year higher and GDP growth 2% higher under such united governments, which can actually enact their programs, than under divided governments, where the President is usually unable to enact much of the program for which he was elected. So, as ever, there is much going on here beneath the water, but things are fascinating, if far from straightforward.