Comments Guidelines

All comments are pre-moderated. No spam, slurs, personal attacks, or foul language will be allowed.

Saturday, April 21, 2012

Greg Mankiw Doesn't Understand Competition for Investment

Greg Mankiw's column in Sunday's New York Times makes the case that competition between governments is a good thing, that it makes them more efficient in the same way that competition among firms does. He paints it as also being about choosing redistributionist policies or not, with Brad DeLong and Harold Pollack both ably making the case that of course governments should engage in redistribution.

As author of Competing for Capital, however, I am more interested in the question of whether government competition for investment leads to more efficient outcomes. The answer, in short, is that it does not. Indeed, competition for investment leads to economic inefficiency, heightened income inequality, and rent-seeking behavior by firms (a further cause of inefficiency).

Mankiw claims:

...competition among governments leads to better governance. In choosing where to live, people can compare public services and taxes. They are attracted to towns that use tax dollars wisely....The argument applies not only to people but also to capital. Because capital is more mobile than labor, competition among governments significantly constrains how capital is taxed. Corporations benefit from various government services, including infrastructure, the protection of property rights and the enforcement of contracts. But if taxes vastly exceed these benefits, businesses can – and often – move to places offering a better mix of tax and services.

Mankiw doesn't stop to think about what this competition looks like in the real world. To attract mobile capital, immobile governments offer a dizzying array of fiscal, financial, and regulatory incentives to companies in sums that have been growing over time for U.S. state and local governments, as I document in Competing for Capital and Investment Incentives and the Global Competition for Capital. His discussion centers on the reduction of corporate income tax rates, which is surely a part of the competition, but which is no longer an issue when an individual firm is negotiating with an individual government.

At that level, the issues then become more concrete: Can we keep our employees' state withholding tax? Can we get out of paying taxes every other company has to pay? Will you give us a cash grant? The list goes on and on. As governments make varying concessions on these issues, you then begin to see the consequences: discrimination among firms (especially to the detriment of small business); overuse and mis-location of capital as subsidies distort investment decisions; a more unequal post-tax, post-subsidy distribution of income than would have existed in the absence of incentive use (a corollary of the fact noted by Mankiw that "capital is more mobile than labor"); and at times the subsidization of environmentally harmful projects. Moreover, many location incentives are actually relocation incentives, paying companies at times over $100 million to move across a state line while staying in the same metropolitan area, with no economic benefit for the region or the country as a whole (Cerner-OnGoal, now in Kansas rather than Kansas City, is a good case in point).

Once upon a time, about 50 years ago in this country, companies made their investment decisions based on their best estimate of the economic case for various locations without requesting subsidies. On the rare occasion when a company did ask for government support, it was at levels that would appear quaint today. For example, when Chrysler built its Belvidere, Illinois, assembly plant in the early 1960s, it asked for the city to run a sewer line out to the facility--and it even lent the city the money to do it.

Today, companies have learned that the site location decision is a great opportunity to extract rents from immobile governments, and invest considerable resources into doing just that. An entire industry has sprung up to take advantage of businesses' informational advantages over governments--and, indeed, intensify that asymmetry--to make rent extraction as effective (not "efficient"!) as possible.

Finally, let's reflect on the force that makes this process happen, capital mobility. The fact that capital has far greater ability to move geographically than labor does, and that governments of course are geographically bound to one place, is a source of power for owners of capital. Modern economists, especially conservatives and libertarians, often have great difficulty acknowledging the role of power in market transactions, though their ostensible hero, Adam Smith, did not. To treat this power as a natural phenomenon rather than a social one, as Mankiw does, is dangerously close to saying that might makes right. But that's not the way things are supposed to work in a democratic society, or a moral one.

Wednesday, April 18, 2012

US takes steps to stop being tax haven

Via @RichardJMurphy, the Tax Justice Network reports that yesterday the IRS and Treasury Department released new regulations requiring banks to report interest income to foreigners' tax authorities. TJN describes this as "a big win for transparency," and indeed it is. It is a welcome step away from the hypocrisy the U.S. has displayed in pressing foreign governments to cooperate with the IRS, while turning a blind eye towards the way the U.S. can function as a tax haven for foreigners.

Predictably, Senator Marco Rubio and Rep. Bill Posey, both Republicans, have introduced legislation that would overturn the regulations, according to Bloomberg. Stay tuned.

Monday, April 16, 2012

The Laffer Curve Refuted

Mike Kimel at Angry Bear has several nice posts on the "Laffer Curve" that underlies much of conservative economic orthodoxy in this country. As you may know, Art Laffer famously claimed that at tax rates of 0 and 100%, you would get zero tax revenue, and that in between, there is an inverted U shaped curve, where taxes collected first increase as the tax rate goes up, then decrease as tax rates go higher still, back down to zero tax collected when the tax rate is 100%.

The Kimel post linked above was prompted by an economist at the American Enterprise Institute, Alan Viard, telling the New York Times that all economists know that when the top tax rate is 35%, cutting rates further will reduce tax revenue.
“The Reagan tax cuts, on the whole, reduced revenue,” he explains. “The Bush tax cuts clearly reduced revenue. There is no dispute among economists about that.”
Except, as Kimel points out, lots of conservative economists dispute this, including one who co-authored a paper with Viard! For his trouble, Kimel became the subject of a post at the AEI blog by James Pethokoukis, which started by completely misidentifying him and going downhill from there. For Kimel's enjoyable takedown of this post, see here.

All this led me back to an earlier post of Kimel's, where he makes an empirical estimate of the Laffer Curve, using U.S. data all the way back to 1929, the first year for which official U.S. data exists. I'll spare you the technical details (see Kimel's post), but here's the bottom line: Laffer got it exactly backward, with tax revenue initially falling as tax rates increase, then rising after a further increase in rates. Here is Kimel's estimate of the "true" Laffer curve:

Not only that, as one of Kimel's commenters, Robert Waldmann points out, we actually have experience with a country having a top marginal rate over 100%, Sweden in the 1970s. Contrary to Laffer, not only was tax revenue not equal to zero, in 1975, Sweden's tax revenue was 41.3% of gross domestic product! (OECD statistics, click on "data by theme," then "public sector, taxation, and market regulation," then "taxation," then "revenue statistics - OECD member countries," then "comparative tables") 21.2% was central government revenue, i.e. excluding subnational government and social security. Either way, a long way from zero.

Not to belabor the point, but Viard was right about tax revenue after President Bush's tax cuts. Here is the OECD data for the federal government, excluding Medicare, Medicaid and Social Security. First we see the effects of President Clinton's tax increase, then President Bush's tax cuts.

Year          Tax/GDP

1992          10.7%
1993          11.0%
1994          11.3%
1995          11.7%
1996          12.2%
1997          12.7%
1998          13.1%
1999          13.2%
2000          13.5%
2001          12.5%
2002          10.4%
2003            9.8%
2004          10.0%
2005          11.2%
2006          11.9%
2007          11.9%
2008          10.4%
2009            8.4%
2010            9.1%

Source: OECD, directions as above for Sweden

Before you supply-siders get too excited about the increase in 2006 and 2007 to 11.9%, remember that the higher Clinton tax rates brought in more revenue for five straight years, 1996-2000.

Though many journalists get it wrong, chessplayers like myself know that "refute" means to conclusively disprove. And the Laffer Curve stands refuted.