At least Americans can agree on one thing. Nothing is more important when going into the voting booth than the economy and jobs.
Just look at the poll results. A recent CNN/ORC poll asked, “Which of the following is the most important issue facing the country today?” The top three responses were the economy, health care, and the federal budget deficit.
“The economy” had more than twice as many votes as health care and the federal budget deficit combined.
Of course, politicians know it and do anything possible to exploit it. Every politician that has stepped in front of a microphone plays up how they are spurring job creation and economic gains for their proposals or slams their opponent for the opposite.
So, if the economy is so important that it dominates elections and politics, is it possible that the elections influence the economy? Yes.
Ignore polls of analysts reporting “where the markets will go”. They are all sell-side cheerleaders.
Also, can you accurately predict the outcome of an election based on the economy? Yes.
Forget all the talking heads and their endless banter about Hillary’s chances.
Statistics are all you need. Let’s take a look at the research…
The Four Year Cycle
First up is a look at how elections affect the economy. It ends up there is a pretty strong link between market returns and the four-year terms of presidents.
Here is a basic breakdown of the four years between presidential terms, using data since 1900:
- The first year sees weak markets and is the generally the worst.
- The second year sees more bear market bottoms occur than in any other and is generally bad.
- The third year is by far the best to be in equities.
- In the fourth year the exuberance dies down. The stock market’s performance tends to be above average, but the cycle is past the peak and heading back to the trough in the new set of four years.
Here is how it looks in chart form:
Of course, the times have changed quite a bit ever since the dawn of the 20th century. The Wall Street Journal did some research on the cycle just using data from 1945 through 2012.
The cycle isn’t quite as dramatic. The low returns in the first and second years aren’t quite as bad. The fourth year is a bit worse too.
The dominance of the third year is still very prominent for returns:
Of course, the markets and the factors that influence them have no reason to neatly line up with any given calendar year. Plus, the election process starts long before a president is sworn into office.
While the election is in full swing in the last year, the markets tend to show more confidence after both major parties confirm their nominees during an election year.
The last seven months of the fourth year of the cycle have delivered positive returns for S&P 500 stockholders in all but three election years since 1945.
What’s more, stock values tend to rise when a sitting president is running for reelection. Since 1900, Dow Jones Industrial Average returns have averaged a healthy 9% whenever the White House occupant wants to stick around.
Predict the Election
So elections have strong correlations to markets, posing an obvious question: “Can market behavior predict who will win?”
Yes it can, and the data backing this up is surprisingly strong.
Sam Stovall of S&P Capital IQ had this to say about it:
“An S&P 500 price rise from July 31 through October 31 traditionally has predicted the reelection of the incumbent person or party, while a price decline during this period has pointed to a replacement. Since 1948, this election-prognostication technique did an excellent job, in our view, recording an 88% accuracy rate in predicting the re-election of the party in power (it failed in 1968). What’s more, it recorded an 86% accuracy rate of identifying when the party in power would be replaced (it failed in 1956).”
Professor Robert Prechter, founder of Elliot Wave International published a study on something very similar:
“[W]e deem an election a landslide victory if the incumbent competed for and won re-election by defeating the nearest competitor with an electoral vote margin of 40% or greater… We define a large positive stock market change as a net gain of 20% or more in the preceding three-year period…
We conclude that a large net positive stock market change during the three years prior to the election is highly likely to be associated with a landslide victory for the incumbent as opposed to a landslide loss.”
Returning to Normal
By now I’m sure you’ve started thinking the obvious. In recent years, this cycle has been off. The housing bubble, credit crunch, and the Fed completely swamped it.
Consider the cycle like waves in the ocean. If a tidal wave comes along, the cause of the normal waves is still there. It just doesn’t have any real impact.
Just look at recent history. When (the second) President Bush entered office the pattern held during the aftermath of the dot com bubble bursting and the 9/11 attacks. In 2003, the markets saw a large gain, followed by a more reasonable gain in the fourth year of Bush’s first term.
His second term bucked the cycle. By the third year (2007), the problems in the economy were becoming apparent. Then the recession hit in 2008 and crushed the markets in late September and early October.
Neither of Obama’s terms have matched the cycle either. The correlation between the election cycle and the markets has been outweighed by extraordinary intervention and interference.
This seems to have pushed the cycle forward, but it should return to normal before the next president enters office in 2016. The Fed is winding down QE and is going to be raising interest rates next year.
If the economy is stable and there aren’t any overwhelming macroeconomic factors, the presidential election cycle creates a predictable pattern in market returns and the outcome of elections.
Forget the market analysts, pundits, and polls. You can do a better job predicting where the markets will go and who will be the next president without them.
