EconLog small logo  

A Recession Soon?

Why Isn't the Military a Stron... Irrational and Negligent...

by Pierre Lemieux

I will not repeat the cliché according to which predicting the future is more difficult than predicting the past, but I see at least three reasons to expect a recession soon.

recession.jpgFirst, the last recession ended nine years ago (in June 2009), which means that the current expansion will have lasted 10 years next month. That long an expansion only happened once before, from March 1991 to March 2001, across the 33 recessions identified by the National Bureau of Economic Research since 1857. Many theories exist about the causes of recessions and they are hotly debated by economists. But as long as the world does not change fundamentally, we must expect recessions to recur with some regularity.

It is true that recessions have become less frequent during the past few decades, but the Great Recession of 2008-2009 was the worst since the Great Depression of 1929-1933. It is also true that economists surveyed by the Wall Street Journal last month put the odds of a recession starting in the next 12 months at only 15%, but macroeconomic forecasts should be taken with a grain of salt, especially when they seem to assume that "this time is different."

Second, interest rates have been increasing for some time, and the Fed wants to push them up further. The Fed's influence on interest rates is limited; at any rate, it cannot be said to literally "fix" them. Jeffrey Hummel and co-blogger Scott Sumner have persuasively argued that point. Whatever the cause of the current and expected upward trend in interest rates, and the more so if the Fed goes against the market, the result could be a crash of stock prices, a rise of bankruptcies, and a general increase of economic uncertainty, all factors typical of recessions.

Third--and this may generate a harder economic shock than the second reason--investors are worried about the smoldering trade war. This fear was already visible in late January, after the Trump administration announced tariffs on solar panels from China and washing machines from South Korea. The Dow Jones Industrial Average (DJIA) quickly dropped by 10% after its all-time peak of January 26. The increase in interest rates was deemed by many analysts to be the main factor, but fears of a trade war certainly played too. Analysts started emphasizing the trade factor after the drop of March 22 and 23: "Trade Fears Jolt Global Asset Prices" ran a Wall Street Journal story. This followed new tariffs on steel and aluminum (mainly hitting friendly countries) and more talks of tariffs and retaliation. The DJIA has recovered some ground since, but it is still lower than in was on January 26 and even on March 21.

Barring an unexpected reversal in the federal government's protectionism, the risk of a worsening trade war persist, even if stock markets may underprice it. As the Wall Street Journal wrote on May 3 ("Dow Industrial Ends Slightly Up After Clawing Back Earlier Losses"),

Trade concerns also continue to linger as the Trump administration has yet to complete implementing plans for its proposed tariffs.
At the very least, a trade war would, like the Smoot-Hawley Tariff of 1930, worsen the recession. At worst, it could fuel a depression.

Other sorts of shock are possible. A constitutional crisis would be one. A geopolitical event would be another one--especially if it pushed up oil prices. A major war would prevent or hide a recession (and is thus, alas, in the interests of some politicians), but it would only give the impression of prosperity.

The main question, I would argue, is not whether a recession is coming, but whether public policy will change it into a depression.

Comments and Sharing

COMMENTS (7 to date)
fwc writes:

What are the chances that Trump is actually trying to spark a recession--or trying to get it here sooner rather than later--with this trade war? Do it early in his term so that the market tanks and he can possibly be credited with pulling the economy out of it just before the next election. The worst year for the market during a presidential term is typically its 3rd year--wouldn't be the first time.

Pierre Lemieux writes:

That's an interesting point, @fwc. Research suggests that voters evaluate the incumbent(s) with the conditions obtaining one or two quarters before the election--see my review of Achen and Bartels. But then, why didn't Trump do this a bit earlier, to be clear (if there is no depression) by the time of the congressional elections, or just after these elections?

Todd Kreider writes:

To Pierre: Maybe a picky point but do you want to say "worsen the trade war" when we aren't in a trade war?


To fwc who wrote: "The worst year for the market during a presidential term is typically its 3rd year-":

I'd say the overall strength of the economy is what most voters care about, not so much the stock market.

Unemployment fell under Carter before shooting up his fourth year, thereby greatly hurting his chances of reelection.

Unemployment fell under Bush Sr. as along with solid growth until his fourth year

Unemployment under Bush was low and growth strong until his fourth year of his second term, thereby making a Republican win nearly impossible.

Pierre Lemieux writes:

Todd: You are basically right, although one could perhaps argue that trade skirmishes have already happened and so the war has started.

aaron writes:

You're assuming the recession actual ended in '09. Really real growth didn't start happening until a fee years ago.

But, yeah, i agree. We are due. And we blew our wad in 09 to paper over the problem without fixing anything. Can the fed even afford to keep rates low any longer?

Jeffrey Rogers HUmmel writes:


I think you make an interesting case, to the extent that anyone can successfully engage in such forecasting. But I do have one quibble. You accept the official NBER’s dating of recessions prior to 1929. But Milton Friedman and Anna Schwartz first pointed out some reservations with the NBER list of early downturns, and then the research of Christina Romer, among others, demonstrated that the NBER dating grossly overstates the volatility of the U.S. economy during the 19th and early 20th century.

I once did an Alt-M post discussing this question. The only genuine U.S. recessions/depressions or major downturns that weren’t war-related prior to the Great Depression are 1920, 1907, 1893, 1873, 1857, 1837, and 1819. Nearly all the gaps between them exceed ten years, and some stretch out to twenty years.

Pierre Lemieux writes:

Thanks, Jeff, for this useful comment. I will be more prudent in the future. I think, however, that you will agree it is still more proper to compare today's economic outlook to the period after the Great Depression.


Return to top