Economic estimates usually imprecise at best

No, the Green Line light rail project did not result in $4.2 billion in economic growth for the Twin Cities.

That someone at the Metropolitan Council recently suggested this shows how hard it is to measure and attribute economic results in the real world and how easy it is to distort such measurements when they are made. The practical difficulties in measurement mean that looking backward in time, it is often hard to tell whether some decision, like building the Green Line, was good or bad. Yet that difficulty pales in comparison to the uncertainties encountered when looking forward.

In fairness, the Met Council did not make any direct connection between the commuter light-rail line and this dollar-figure increase in economic activity. The Council merely added up the value of new construction within a half-mile radius of the line, including the end points in both downtowns. For downtown St. Paul, this means that even some new construction south of the Mississippi gets included. Yet the figure, and the connection, was reported — and distorted. The mistaken assertion I overheard was in a coffee shop discussion of the news.

The Met Council obviously has an agenda in the matter of rapid transit, with the funding of an additional line pending in the Legislature. So skeptics — or cynics — who question the political motivation, as well as the methodology, behind the $4.2 billion number may have some grounds. But we’ve all seen this before: The U.S. Army Corps of Engineers tends to find data that show its flood control and navigation projects turn out rosy; the Bureau of Reclamation finds great recreational and other ancillary benefits from its dams. Thus, some skepticism is warranted.

And while this specific number is questionable, the Green Line certainly has spurred economic activity that would not have occurred in its absence. It also stifled some, especially during construction. So while it is a mistake to overestimate net beneficial outcomes of public investment, it is equally erroneous to underestimate them. Unfortunately, coming up with really trustworthy numbers often is difficult and sometimes impossible.

In academics, inferring that a rail line caused development is a post hoc ergo propter hoc fallacy, which freshmen economics or philosophy students are strictly warned against. But such “A happened and then B happened, so A caused B” logic dominates public discourse, especially in election years. For many people, the sequence — and consequence — of two events is proof of cause and effect.

So how does one get beyond this?

Even the Green Line’s strongest advocates acknowledge that while some development within an arbitrary radius was directly caused by the line, it was only one factor among many for other investment decisions. For some projects, such as that at Hennepin County Medical Center, the existence of the line didn’t play any role.

A key question, therefore, is this: How much activity would have taken place had the line not been built? One approach would be to put things in historical context. What was the annual average of such investment, adjusted for inflation, in these neighborhoods before the line became a factor? What was the trend over time in this spending? How far are current levels or trends above those of the past?

One could also try to compare investment along the line with that in other areas where transit is not a factor. Break the zone up into various categories and identify discrete neighborhoods or sectors — residential, industrial, retail etc. Find similar zones elsewhere, pair them and compare levels of new investment.

This is what an appraiser recently did with 80 acres of farmland I am selling. There were no comparable complete farms sold in the past year within several miles. But one can break my 80 acres into tracts such as excellent, average and poor crop land, good pasture, wasteland and so on. The same can be done for other farms that, taken as a whole, are very different from mine. The total sale value of these other farms can be broken down into component tracts and those values applied to similar ones on mine. Some weighted value of the whole 80 acres is then tabulated.

This is highly useful but still involves much subjectivity. Which brings us to the larger problem of looking forward, using subjective cause-and-effect methodology to justify spending now to achieve returns later, as the Met Council seemingly would have the Legislature do with future rail projects.

In many sciences, organized experiments to determine outcomes are a key method. But such organization is difficult in real world economic estimates. As in appraisals, comparing results of different transactions, historical trend versus cross-sectional, helps. However, much uncertainty remains.

In these economic situations, one essentially must answer Dear Abby’s classic question: “Are you better off with him or without him?” Republican presidential candidate Donald Trump repeatedly frames future defense policy in financial terms. He asserts that our country pays much more than its fair share of NATO costs. The issue is an old one.

As with the cost and benefits of Minnesota rail lines, economists would approach the NATO issue in terms of “opportunity cost.” How would U.S. Treasury outlays change if we no longer spent as much on NATO? That is a hard question to answer.

Critics often take the total costs of U.S. troops stationed in Europe or South Korea as a direct expense to U.S. taxpayers. During the Cold War, our four army divisions and several air wings in Europe cost a significant percentage of defense outlays. But if we did not have those units overseas, would we have simply spent the money elsewhere? Or did we see some tangible result, beyond monetary, of our military commitments?

Trump argues that a businessman of his experience can jawbone other nations into paying. Perhaps. But the only threat he can offer is U.S. withdrawal. The credibility of that threat depends on the change in U.S. outlays versus the change in the value of the level of defense achieved. Either way, under a future President Trump, even if our alliance partners call his bluff, we are still paying less money on this part of our defense. Would the consequences of such a change be beneficial in the long run? This is an economic question for which specific dollar estimates are difficult to assign.

That is the dirty secret of many economic studies, as with the boiler-plated disclaimers on investment plans — “past results are not an indicator of future returns.” Economic theory and history do give us some idea of the probable direction for the results of some economic choice. But estimating a specific dollar value of the outcome, after the fact as well as before, usually is maddeningly difficult and often impossible. We can camouflage uncertainty with financial mumbo-jumbo, but many real-world decisions, whether to buy a farm, build a transit line or leave a defense pact, ultimately get made on very subjective bases.