Saturday, July 22, 2017

Share Share    Print Print    Email Email

New Jersey’s “Stretched” Jobs Numbers from Another Dimension

The New York Post reports that New Jersey Governor Corzine’s office suggested Cabinet members find him speaking events that show job creation or economic development in the private sector, writing in an email, “I know that it might be a stretch for some of you, but please be creative.”

Rather than find fault with political actors acting (unsurprisingly) in their own self-interest, it’s bad policy resting on weak theory that lends itself to creativity in economic calculation.

My Mercatus colleague Veronique de Rugy explains what this fuzzy stimulus-math has to do with the job creation multiplier, a Keynesian-inspired calculation that sits on shaky theory.

First the theory. Keynes asserted that during downturns consumers hoard money. Thus, less is spent in the economy. The policy solution was for government to spend to spur economic activity. In 1930, Richard Kahn built on idea, creating the “Keynesian multiplier,” demonstrating how a government dollar stimulates more spending in the private economy.

Simply put, when the government spends $100 million on workers to pave a road, the workers spend that money in different businesses, which now have more money to spend — i.e., hire more people, who will now have income to spend. Sounds simple. But the theoretical problems are many:

  1. In his model, Keynes separated saving from investment, a fatal flaw. Saving is not necessarily money hoarded, it is money put in the bank and lent out; that is, invested.
  2. Keynes’ model is based on an “aggregate consumer” which abstracts away from the reality of an economy comprised of individuals all facing their own unique circumstances.
  3. The government cannot create wealth, it can only redistribute it from the private sector. At the other end of government stimulus are taxes.

Then there is the problem of calculating the effects of government spending on the economy. When wading into multiplier territory, it can get technical fast. Here’s a discussion of different approaches.

Carl Bialik of the the Wall Street Journal notes that jobs measures differ between states and the CBO due to wildly different assumptions and huge counterfactual leaps:

In every method used, economists are forced to imagine an alternate reality — one built on assumptions that are easily challenged. For example, to compare present unemployment rates to past rates may be straightforward but it fails to account for other economic forces that were going to affect unemployment with or without the stimulus.

Given the “alternative reality” problem, is there really much difference between jobs figures stretched for political effect, and those generated via sophisticated models? Depends on what parallel universe you like to live in.