“Rich States, Poor States, 6th Edition”

Arthur Laffer, Stephen Moore and Jonathan Williams write:

All across the nation, states are looking for ways to boost their economies and become more economically competitive. Each state confronts this task with a set of policy decisions unique to their own situation. . . .

Fortunately, the United States, with its “50 laboratories of democracy,” provides us with empirical evidence to track exactly [sic] which policies lead to economic prosperity and which fail to deliver. . . .

Armed with years of economic data and empirical evidence from each state, the authors identify which policies can truly [sic] lead a state to economic prosperity. Rich States, Poor States not only identifies these policies but also makes sound research-based conclusions about which states are poised to achieve greater economic prosperity and those that are stuck on the path to a lackluster economy.

I’m supportive of this sort of effort but I don’t think the authors help their credibility by stating their conclusions with such certainty. Can’t the American Legislative Exchange Council hire an economist or political scientist or statistician who can tell them about causal identification?

The report is pretty long and I did not read the whole thing. But it looks like what they’re doing is giving each state two rankings:

– An “economic performance rank” based on averaging the state’s ranking on three measures: growth in state GDP (in percentage terms, but not per-capita, for reasons they discuss in their report), net domestic migration (this time in absolute numbers, not percentage), and non-farm payroll employment (in percentage terms). I don’t know why the only include domestic migration, nor do I know why they exclude farm employment; the explanations are somewhere in the report, I assume.

– A policy measure (they call it an “economic outlook rank,” a “forward-looking forecast,” but it is actually a measure of existing policies), that’s an average of the state’s ranking on 15 categories, many of which involve tax rates on business and rich people. (Items 1, 2, 3, 4, 7, 11, 12, 13, 14 directly or disproportionately affect businesses and the rich; items 5 and 6 affect the general population more or less equally, I think; and I’m not so clear on 8, 9, 10, and 15.)

Their claim is that these two measures are highly correlated and that the second measure is predictive of the first. That is, reducing business regulation and cutting taxes on the rich will lead to absolute economic growth.

Or, to give it a more political spin: conservative economic policies are good for the economy, liberal economic policies are bad.

This conclusion definitely could be correct, especially in the long term. Here’s what I wrote about an earlier edition of the Laffer et al. report:

In chapter 5 of Red State, Blue State, we pointed out some patterns that might be relevant to the point made by Laffer et al. One issue is that economic performance can be defined in different ways. For example, from 1981-2004, the five richest states in per-capita income (Connecticut, New Jersey, Massachusetts, New York, and Maryland) performed poorly if you look at incomes of the 10th percentile within each state, but they performed well for the 90th percentile.


In poor states, the poor have been doing better; in rich states, the rich have been doing better.

So it is plausible to me that the historical analysis done by Laffer et al. would show conservative policies being correlated with economic growth. They just have to ease off a bit on their causal claims, given the obvious issues of identification (really, it’s too bad they didn’t have an economist or political scientist or statistician involved in some way in the preparation of this report—-we can be hired for pretty, you know!). Descriptive analysis is just fine, I do it all the time.

Also, in some places I don’t think the authors have fully worked through the implications of their arguments. For example, #1 of their “Golden Rules of Effective Taxation” is “When you tax something more you get less of it, and when you tax something less you get more of it.” But on pages 6-7, under the heading, “Tobacco Taxes Threaten the New Hampshire Advantage,” they recommend that this state continue a plan of tax competition, keeping cigarette taxes lower than that of neighboring states so that people will drive to N.H. for their smokes. Lowering cigarette taxes so people smoke more, that hardly seems like “effective taxation.” I mean, sure, there are some people who would support an increase in smoking, but this is clearly a minority position.

Finally, I was curious about this group’s stand on social issues so I went their website and searched “abortion” but all I got was this press release which didn’t give their stand on the issue. I directly googled alec abortion and it appears that the organization used to be strongly anti-abortion and focused on social issues, but maybe that’s not on their agenda anymore. I recalled that this organization was active recently on guns, but searches of “guns” and “rifles” on the ALEC website yielded no hits. Perhaps they’ve ditched the social issues after all that bad publicity, I don’t know.

P.S. The book even includes a picture of the famous Laffer curve! So retro. I was expecting to see some shots of bell-bottoms and Hawaiian shirts, maybe an REO soundtrack for the whole thing?

7 Responses to “Rich States, Poor States, 6th Edition”

  1. Funkhauser May 27, 2013 at 10:42 am #

    Thank goodness nothing that happens in California affects outcomes in Nevada. Dodged that bullet, they did.

  2. jonathan May 27, 2013 at 11:33 am #

    My favorite ranking is the donor/donee states ranking. Not because it has some special value but for the way it reveals a series of hidden ideas. Such as:

    1. The donee states are nearly all run by the GOP. That leads to the obvious point that the people so much against government in fact receive more than they pay in taxes. Less obvious is that within the donee states the belief seems to be – note the seems – that “others”, meaning poor and minorities, receive these funds. This suggests a bunch of things, including whether people within these states believe the excess funds (and mandated spending by their own state and local government) somehow causes “issues” related to the poor and minorities.

    2. The excess federal money is a subsidy that allows these states to maintain policies which the donor states don’t like. The donee states, perhaps in part because of ancient regional animosity, spend much energy using tax subsidies and other devices to take businesses away from donor states. The donee states maintain regimes of lower taxation and regulation in part because they receive excess funding from donor states that allow them to avoid raising taxes to pay for the social programs other states may substantially fund. And so on.

    So using the “conservative” approach, I could argue that being a donor state is a mark of failure, that the state’s own economic system is not capable of addressing issues and thus it relies on federal money as a crutch. I admit some of my thinking on this is anecdotal; when I’ve raised the issue with white southerners, I hear that they receive more money because they have more poor people – and lots of minorities, meaning African-Americans. (I note in response that much of the world has raised itself out of poverty since the Civil Rights Era in the US and thus I question the actual efficacy of the economic system of these laggard states.)

  3. Thomas May 27, 2013 at 1:12 pm #

    I think you’ve been misled by the poorly formatted webpage. My reading is that the ALEC website is announcing the publication of the report authored by Laffer, Moore and Williams, not that the website notice itself has been authored by them. Your particular complaints about the certainty with which the authors have stated their conclusions in that notice then is almost certainly not about anything the authors themselves have stated.

    • Andrew Gelman May 27, 2013 at 6:51 pm #

      Yeah, they should find some better-quality blurb-writers.

  4. Sebastian May 27, 2013 at 9:10 pm #

    There is an old joke about this group of scientists that get asked by a commission what 2+2 is. You get all the predictable stuff – mathematician says it’s trivial, engineer says it’s 3.9993 etc. and then the economist comes in, closes the door behind him, draws the curtains and says: “well, it depends on the assumptions. What do you _want_ it to be?”

    Point being that adding an economist (or a sympathetic political scientist or statistician – maybe Wegman is looking for a new consulting gig?) wouldn’t necessarily help if you’re looking to get a certain result. And if you’re hiring Laffer (who is, technically, an economist – he’s got a Stanford PhD in econ), you do want a specific result.

  5. Mark May 27, 2013 at 10:49 pm #

    I have trouble with the basic thesis underlying the study that we have “50 laboratories” that allow studies to be done. If one were to do an experiment with 50 subjects one would start with subjects that were as identical as possible, start them all from an identical baseline, and randomly assign them to the intervention (often with blinding of the experimenter.) But the states are quite different from one another and have their own histories, politics and traditions. And obviously, changing their freedom milieu to study its impact can’t be done.
    The definition of freedom is open to question. North Dakota famously banned “Slaughterhouse Five” but is commonly listed among the free states. I know that it has been some time since the ban, but culture tends to persist.
    I have an alternative hypothesis. I think that states involved in hydrocarbon extraction are doing well. This explains many of the free US states and also works for Alberta, Norway, and Saudi Arabia.

  6. Craig May 28, 2013 at 11:12 pm #

    From his choice of words it seems that Prof. Gelman has already suspects the slant here. The choice of which variables to measure probably determines the outcome. Select three different economic measures or a different smorgasbord of tax policies and you’ll almost certainly get different results. Cynically, one could guess that the goal of the study might be a Reinhart-Rogoff headline binge followed by years of citation by the faithful. The years chosen for study and so on probably have impacts as well.

    More interesting question: why is one’s initial assumption upon seeing a GOP economic headline more or less: “GOP economist(s) gaming numbers again”. To an outside observer, GOP economics seems increasingly like a fever-dream, or like the tobacco company scientific claims of 1970, 1980 or 1990, or climate change denial.