Art Carden  

Government Projects and Subsidies: What are the Alternatives?

PRINT
Teaching the Children by Examp... John Goodman on Why We Have Po...

Repeat after the economics profession: resources are scarce, and they have alternative uses. Thomas Sowell has said that this is the first rule of economics. He has also said that the first rule of politics is to ignore the first rule of economics, and this is perhaps nowhere more obvious than in discussions of state and local development policy.

The state of Alabama has given hundreds of millions of dollars in subsidies and tax breaks to auto manufacturers, and the city of Birmingham has been talking for a long time about building a domed stadium and expanding the convention center. I've seen $500 million listed as a price tag for this venture, but I don't know that the city's prospective commitment is that high.

"What else could we do with the money?" is the question too few people are asking. Questions about economic calculation are important, but given that governments are doing these things in the context of a market economy we can at least use a few benchmarks.

So how should governments evaluate their undertakings? Ignore public choice considerations for just a second and indulge a flight of fancy. There is a collective action problem that, in theory, could mean too few stadiums and the like get produced and that could, in theory, mean that government provision of stadiums and the like would make us all better off. If Alabama, for example, is a better place to live because a government spent $250 million to lure CarCo or to build a stadium, the indirect benefits should be reflected in higher real estate prices and, therefore, higher property tax revenues. The "intangible benefits" of "putting Alabama on the map" or "becoming a big-league town" are more tangible than they might at first appear because they will be capitalized into real estate prices. Hence, we could estimate the project's contribution to tax revenue in order to determine whether it's actually creating value.

Of course, there are a lot of ways to use $250 million. A government could fund a stadium, give it to a car company, cut taxes, pay for better schools, or simply invest it in stocks and bonds. What is the baseline to which we should compare government spending on economic development, and how should we evaluate the outcomes?

I'm tempted to say we should evaluate taxes and spending by comparing it to a program of investing in stocks, bonds, or a combination of the two, distributing the proceeds, and relying on entrepreneurs to provide the things governments currently provide. Essentially, we turn states into big mutual funds. Stocks, though, are too risky while risk-free bonds are too conservative. What benchmarks and metrics would you propose?


Comments and Sharing


CATEGORIES: Public Goods




COMMENTS (6 to date)
Phil writes:

The rate of return on stocks is no less risky than the return on the project the government is considering providing. Therefore, I would argue, the return on stocks is an appropriate benchmark.

Suppose Acme, Inc. is created to (privately) build and own the stadium. Isn't the rate-of-return risk of Acme Inc., stock EXACTLY the same as the rate-of-return risk if the government built the stadium itself?

If so, why assume that Acme, Inc. stock is less risky than, say, Coca-Cola stock?

ThomasH writes:

This really is not that difficult. Just use standard cost benefit analysis.

Do the benefits from the stadium minus the costs when discounted at the rate at which Birmingham can borrow exceed zero? Now of course the terms of the deal (rent of the stadium?) still have to ensure that the government captures the benefits and if the renters balk, it may be a sign that the benefits are mis-estimated. :)

At least that's how Liberals would analyze things.

[I'm assuming that the stadium is evaluated as an investment, not collective consumption like a homeless shelter where capture of the benefits by taxpayers are not appropriate.]

LD Bottorff writes:

The problem with measuring benefits via real-estate value is ensuring that the only cause of increase of the value was the stadium. How do you know that? Also, how far from the city do you count the benefits? For a stadium in Birmingham, will you just count real-estate in Birmingham, or all of Jefferson County? How about real-estate values in Shelby County? Here in Louisville, should we count real-estate values across the river in Indiana? (much of southern Indiana is actually closer to the Yum Center than people in the east end or southwest side of Louisville)

The reason politicians can get away with spending money on these types of projects is that it is so difficult to measure the benefits, so politicians get away with saying that the benefits more than cover the costs.

ThomasH writes:

LD Bottorof's suggestion that stadium benefits might be measured by increases in real estate values suggests a funding mechanism for paying for the bonds, a differential property tax surcharge on property which according to the benefits model will benefit from the stadium until the increases are realized and incorporated into the tax base.

Thomas Sewell writes:

Total expenditure properly time discounted / People in community = Cost, AKA lost benefit per person

A Simple Cost/Benefit analysis would be to take that cost per person and say, is this project worth more than taking that same money and giving it out to all those people to spend on whatever they want? Would the vast majority of them choose to spend it on this project?

Most of the time, the answer will be no, because people will individually spend that money (even if it's only $50) on something more personally important to them than the proposed project.

That's why governments should have a specific limited set of things to spend money on and if they ever want to expand that list, should at least be required to have a super-majority of the people who will be required to pay for it directly approve it.

That's not a perfect solution, but it's at least progress in the right direction.

J Scheppers writes:

The more I see of cost benefit analysis the more I feel these have way too much bias to be appropriate. The most recent trend I see is claims that consumer surplus on the benefits side should be claimed, but no consideration of the loss of consumer surplus on the taxed (subsidy provider) side is every considered in the calculation.

My rule is that there are such things as positive and negative externalities, however those elements if not valued with revealed preference should never be more than 20% of the benefits. Even when revealed preference data is available non-monetary benefits should be no more than 50% of the cost and at least 50% of the funding should be cold hard cash from the facility users minimum.

Randal O'Toole laid out some good principles that should be applied on his blog this week:

Economics for Planners Part 1
http://ti.org/antiplanner/?p=9716

Economics for Planners Part 2
http://ti.org/antiplanner/?p=9722

Never ever believe there is a benefit if the advocate is the same as the beneficiary and they don't want to put any money into the pot. The corollary: The voracity of a claim for a need should be measured by the money the advocate is willing to spend making it a reality.

If the advocate paid for the B/C study it is more likely marketing material than scientific research.

Comments for this entry have been closed
Return to top