Why Are the Forecasts So Wrong?

At Discourse Veronique de Rugy and Jack Salmon point out that many of the economic forecasts have been amazingly wrong:

Since many economists failed to remember the lessons of the past, they continually made bad predictions this year. One of the biggest culprits was Nobel Prize winner and New York Times columnist Paul Krugman, who in August claimed that the expiration of the $600 a week in extra unemployment compensation in late July would drive down consumer spending and result in a 4% to 5% contraction in gross domestic product. Since then, personal consumption has, in fact, continued rising, recovering 91% from its April lows, while the third-quarter GDP expanded 7.4% from the second quarter.

Back in July, economists from Jared Bernstein, tapped by President-elect Biden to serve on the Council of Economic Advisers, to Ernie Tedeschi, a former senior adviser at the U.S. Treasury, forecast that by the end of the year the economy will have contracted by 2% and the unemployment rate would probably be greater than 11%. Similarly, economists at the Economic Policy Institute proclaimed that the end of the extra jobless benefits would lead to the loss of more than 5 million jobs, while Mark Zandi of Moody’s Investors Service warned of large job losses and double-digit unemployment rates “well after the pandemic is over.”

In reality, the economy has continued to expand, employers have added more than 6 million people to their payrolls since July, and the unemployment rate fell to 6.7% last month.

All of this is extremely easy to explain. First, try as they might, economists have not been able to transform economics into a predictive science like physics or chemistry. Flipping it around, if chemistry were like economics, every chemical plant in the world would be exploding unpredictably and frequently and if physics were like economics sending a payload into orbit would be just as likely to hit Tokyo. As John Kenneth Galbraith wisecracked years ago, the purpose of economic forecasting is to make astrology look respectable.

The second reason, as I continuously harp, is that economists are human beings with preferences, beliefs, agendas, and prejudices and all too frequently those overwhelm their professional judgment.

Neither of these deficiencies should be construed as proving that economics is useless. Economics has one fundamental observation: incentives matter. Nearly everything else derives from that observation. The reason that economics remains largely descriptive rather than predictive is that while incentives matter they don’t matter deterministically but vary in degree and vary from individual to individual.

My solution is not to limit my reading to Paul Krugman, Jared Bernstein, or Mark Zandi but also to read the observations of Republican economists like Greg Mankiw or John B. Taylor and libertarians like Tyler Cowen or Scott Sumner, even MMTers like Warren Mosler or Stephanie Kelton.