More Evidence Against Keynesian Economics

by Dan Mitchell | Aug 4, 2026

Keynesian economics assumes government can jump-start the economy by having the government borrow and spend, either directly or by giving people handouts.

According to Keynesian theory, it doesn’t matter how money gets spent. Even if something bad is the reason, like a terrorist attack or natural disaster.

I’m not joking. Here are examples of how various people have justified Keynesian stimulus schemes.

Perhaps the most shocking example of misguided Keynesian analysis occurred in 2001.

Here are some excerpts from a column by Paul Krugman, which was published by the New York Times just three days after the 9-11 terror attack.

…the terror attack…could even do some economic good. …If people rush out to buy bottled water and canned goods, that will actually boost the economy. …the destruction…will generate at least some increase in business spending. …Now it seems that we will indeed get a quick burst of public spending, however tragic the reasons.

I explained why Keynesian theory is wrong in a 2008 video, pointing out that government can’t inject money into the economy (the spending) without first taking money out of the economy (the borrowing).

But I wasn’t saying anything original. Frederic Bastiat actually debunked Keynesianism back in the 1800s.

And we have reams of evidence that Keynesian economics fails whenever and wherever it is tried.

Yet the idea never dies, probably because it is very convenient for politicians (it tells them their vice – reckless spending – is a virtue).

But I’m going to try, once again, to put a final nail in the Keynesian coffin.

Here are some excerpts from a new study published by the International Monetary Fund. Written by Ha Nguyen, Mehdi Raissi, Bruno Versailles, and Alice Tianbo Zhang, it examines whether natural disasters are good for growth or bad for growth.

This paper provides new evidence on the cross-country growth effects of natural disasters using high-frequency climate anomalies and carefully calibrated physical thresholds to identify disasters for a global panel of 196 countries spanning over five decades. We find that storms, floods, droughts, and heatwaves significantly reduce GDP growth contemporaneously by approximately 0.1–0.2 percentage points on average… Severe disasters impose far larger costs. For example, a catastrophic flood can lower growth by up to 3 percentage points, while once-in-100-year storms or heatwaves reduce growth by around 0.5pp and extreme droughts by about 1pp.

Here’s Figure 1 from the study, showing that natural disasters don’t produce more growth.

Simply stated, the Keynesians are wrong.

As Thomas Sowell explained, government spending is a sedative rather than a stimulus.

P.S. If you want to enjoy some cartoons about Keynesian economics, click here, here, and here. On a more serious note (but still entertaining), I highly recommend the famous video showing the Keynes v. Hayek rap contest, followed by the equally enjoyable sequel, which features a boxing match between Keynes and Hayek.