According to Beneficiaries, All Government Spending Is Worth the Investment

Government spenders use a common playbook to convince policymakers to open the public purse. In it, the multiplier analysis is the most commonly called play.

Residents expect demonstrable economic results from policymakers — especially with a 24-hour news cycle to account for their policies’ productivity. Thus, every state legislator is accountable to monthly employment statistics. Monthly employment fluctuations, however, are rarely caused solely by policymakers' actions, but rather by the complex interplay of economic forces.

Regardless, Lansing insiders hold policymakers to a higher standard: proof is required that a politician’s policies pushed the economic needle. This environment creates a market for abusing the multiplier analysis.

These analyses are manipulative because every economic action creates secondary and tertiary consequences. Your window gets broken so you buy a new one — your contractor makes some money, but maybe you don’t put as much on your next grocery list. This continues ad infinitum as you consider the contractor, the grocer, and all the people who would follow after them. Actions wind up encouraging and discouraging consumption, investment and savings well beyond their initial intent. You simply can’t drop a dollar on the sidewalk without it making an echo somewhere else.

Using the multiplier analysis like this creates synthetic statistics — they report economic activity as though there’s a tracking device on compensation, which is of course, implausible.

Sometimes the multiplier analyses look at job creation while others analyses look at government or private-sector spending. Here is a table of multipliers that have been used to justify Michigan spending and tax credits:

Many analyses suffer from the garbage-in, garbage-out problem. An infamous example would be the RASCO fiasco, which looked at an employment multiplier. The state awarded a refundable tax credit to a convicted embezzler who had promised to develop water and electrical modules for use by third-world countries. The MEDC’s analysis showed that there were going to be 1,048 jobs created in the surrounding area because of the project, which was used to justify state support. But the project was a scam; no jobs were created, and fortunately the state repealed its incentives. Multiplier analyses are inherently flawed when there is no project from which to generate the numbers.

While the results of multiplier analyses are the same, the methodologies are inconsistent from one study to another — sometimes to the point where the analysis isn't good for anything.

Earlier this year, a group attempted to show the impact of the state’s art subsidy program by counting the spending at the organizations that it supports. This created a times-51 spending multiplier, meaning that $1 in arts spending generates $51 of economic activity.

Thus, it acts as if the Ann Arbor Film Festival would only exist with subsidy, and it certainly has proved otherwise. The festival occurred without state support, indicating that the assumptions of the times-51 spending multiplier are faulty.

Despite these mathematical shortcomings, manufactured political effects such as these are often enough to warrant repeating in the media. Most people understand that spending $1 has an economic ripple effect, so any judgment from experts seems plausible.

As economist (and former Mackinac Center intern) Daniel J. Smith remarked, “The increasingly sophisticated reiterations of old Keynesian models still fail to pass the inspection of basic economics, and still fail to address the most basic question, ‘where is the money coming from?’” (Feel free to check out a paper he co-authored on the subject.)

All things being equal (and they never are), multipliers can show lawmakers which spending area will have the most economic impact, but it does not justify taking the money in the first place.

Warped figures do taxpayers interested in real accountability a disservice when they are abused to justify takings.