Noah Smith and Matthew Yglesias on Conservative Bias.

Both Noah Smith and Matthew Yglesias have weighed in on my earlier post in which I speculated that one reason that economics faculty are sharply more conservative relative to faculty in other departments on college campuses is that the facts and analysis in economics tend to lead to somewhat more conservative interpretations of the world.  The posts by Noah and Matthew both have good points and I wanted to comment on them briefly before moving on.

First, I should point out that my original post was really about the different political viewpoints of college professors, not typical members of the Democratic or Republican Party.  Both Noah and Matthew point out correctly that the facts actually roughly agree with what an informed policy analyst would believe (either a moderate Democrat or moderate Republican).  Despite the usual public disputes between Republicans and Democrats, better informed members of these parties (e.g., people like Peter Orszag and Douglas Holtz-Eaken, or even Paul Krugman and Greg Mankiw) would actually have surprisingly similar policy assessments.  The range of differences in policy recommendations among mainstream economists is overall pretty narrow. This is also probably due to the disciplining nature of quantitative facts.

In Noah’s response, he says that in the 1980’s,

conservatives moved away from the facts because they could, and liberals moved closer to the facts because they had to.

I think this is correct though the question of why this occurred is left open.  It’s possible that the electorate simply became more conservative and this forced both parties to realign.  Alternatively, it could be that the accumulation of facts may have forced liberals to adopt a more conservative tone as our understanding of the world was refined.  Republicans in contrast could afford to adopt a less realistic view that was even further to the right. I thought initially that Noah’s quote was more like the latter view but I’m not sure…

Both Noah and Matthew suggest that there are many facts which fit better with modern liberal agendas.  Matthew mentions a variety of standard market failures that are typically highlighted in “Econ101” texts.  I’m not sure I entirely follow his line of reasoning – my original reaction was that  it seems like he’s agreeing with Noah’s statement above.  That is, the modern liberal approach to economic policy is to start with a basic free-market approach and then perhaps use corrective (Pigouvian) taxation to improve the allocations as needed (is Matthew channeling his inner Greg Mankiw?).  You won’t find many economists who would take issue with this view but at the same time this seems like he is conceding the point — start with Milton Friedman and then season lightly with a little bit of Paul Krugman?

Noah points out that because many of the incentive effects are small that liberals are tempted to ignore them as we pursue our goals.  The problem with this argument is that it applies equally well to jelly doughnuts and cigarettes.  One jelly doughnut, or one cigarette, does no appreciable health damage.  This doesn’t mean that you can eat jelly doughnuts (or smoke cigarettes) without worrying about their long-term health consequences.  If you pile on tax after tax, impediment after impediment and so on, then the costs will end up being relatively high even if the costs associated with any one regulation are small.

Noah also points out that wealthier nations typically have greater fractions of GDP allocated by the government.  The implication, I guess, is that this is a sign that more government involvement can’t be that bad if the wealthier countries are opting for it.  I don’t think I agree with this statement.  It’s true that most European nations have greater government involvement in their economies but it is also true that, on the whole, these nations are not as productive as the U.S.  Many prominent European nations (e.g., UK, Italy, France, Germany, Spain) have per capita levels of GDP which are roughly 65 – 75 percent of U.S. per capita GDP.  This is a comparison of average GDP per capita.  I would anticipate that median GDP per capita is much closer (since income distributions are much more balanced in Europe on average), but even then, these economies are starting out with much less average output per person to allocate (these measures include government provided goods and services).

I would encourage Noah and Matthew (and anyone else who’s interested) to take a look at the papers by Djankov and Shleifer (and many coauthors) linked below.*  These papers present an empirical analysis of economic performance across nations together with measures of corporate taxation, labor market regulation and regulation of business entry.  Of course, making such a comparison is difficult – there is no sense in which we can take these measures as causal.  Even more problematic, there are many variables that are changing simultaneously.  There are hundreds of policy differences, cultural differences, etc.  This will make interpreting the results extremely difficult.  Nevertheless, there are noteworthy patterns in this data and the basic message is not encouraging.

From the abstract of the corporate tax paper:

[High corporate tax rates have ] a large adverse impact on aggregate investment, FDI, and entrepreneurial activity. … Corporate tax rates are also negatively correlated with growth, and positively correlated with the size of the informal economy.

From the abstract of the paper on regulation of business entry:

Countries with heavier regulation of entry have higher corruption and larger unofficial economies, but not better quality of public or private goods. Countries with more democratic and limited governments have fewer entry regulations.

From the abstract of the paper on labor regulation:

richer countries regulate labor less than poorer countries do, although they have more generous social security systems. The political power of the left is associated with more stringent labor regulations and more generous social security systems. … Heavier regulation of labor is associated with a larger unofficial economy, lower labor force participation, and higher unemployment, especially of the young.

Of course, many European nations are trying to undo some of the policies they have adopted over the years because they are looking at the same data we are looking at. Unfortunately, we don’t really know which particular policy (or policies) is to blame for resulting in a stagnant labor markets but this is little comfort to a nation if it has low labor force participation, low productivity and high unemployment.  Hopefully this more mature liberalism that Noah argues now characterizes the center left in the U.S. has learned from the experience of its forerunners.

*These papers are a bit out of date. The most recent one was from 2008 but the earliest one dates back to 2000.  There are probably more up to date papers looking at these topics.  If there is an expert who wants to weigh in I would be interested in hearing about more current work.


27 thoughts on “Noah Smith and Matthew Yglesias on Conservative Bias.

  1. Many prominent European nations (e.g., UK, Italy, France, Germany, Spain) have per capita levels of GDP which are roughly 65 – 75 percent of U.S. per capita GDP.

    They may be heavily weighed down by poorer, more people-dense countrysides. Great Britain’s GDP per capita is 77% that of the US, but London’s GDP per capita is higher than that of New York City. The US historically had and has relatively high farmer incomes outside of the post-slavery South.

    • In addition, many European countries have far lower economies of scale than the USA. The US has had (guessing) as many people as Western Europe for quite a while, all speaking one language, with much closer to a single legal system. Throw in the facts that the USA (1) has avoided a major war on it’s own terrain for a century and a half, (2) gained enormous wealth through taking enormous swathes of thinly-occupied land and (3) was able to exploit much of that by a large system of slave (and later serf) labor.

  2. I agree with Brett, there are many reasons why there would be a large income difference between (Western) Europe and the United States. If we are to compare the effect of government size on gdp per capita, it surely makes more sense to make a comparison between countries within europe.

    The way you percieve the political divide seems a bit dated. To quote:
    “start with Milton Friedman and then season lightly with a little bit of Paul Krugman”
    That sounds very much like what todays democrat party wants. If an economist with the views of MIlton Friedman were to make an entrance into todays public debate, he would be considered mildly left-wing.

    To take an example, the kind of healthcare reform that Reagan was trying to pass involved at least as much state intervention as Obamacare does, probably more.

    That said, I enjoyed the column! Especially the part about how regulations that look good in isolation can be very harmful put together.

  3. “If an economist with the views of MIlton Friedman were to make an entrance into todays public debate, he would be considered mildly left-wing.”

    I’m not sure I agree with HA’s point: the vast majority of research in macroeconomics, especially over the past twenty years, has been aimed at analyzing the effect of incomplete, non-competitive, or otherwise imperfect markets. If economists hold the view of Friedman at his “purest” it is they who’d look “dated.”

    More recent work on distortions and income levels (not just US vs. EU) has been done by: Nick Bloom and co-authors, Diego Restuccia and Richard Rogerson, and importantly, earlier work in the 90s by Steve Parente and….. Ed Prescott. (and you have to read the papers–it’s not useful to say “bah, ed prescott!”–After all, he put a model down, parameterized it, and gives us numbers. If we don’t like it, we have write something better down.)

  4. Chris: “… (e.g., people like Peter Orszag and Douglas Holtz-Eaken, or even Paul Krugman and Greg Mankiw) …”

    Considering that DeLong and Krugman have covered rather, ah – problematic behavior by Douglas Holtz-Eaken, and Mankiw is sorta famous for flipping quite solidly before, during and after his tour in the Bush administration, perhaps some proof of this is in order.

  5. Krugman has written about his personal journey from Econ Prof to political commentator and noted how wrong many of his background assumptions about the actual practice of politics were. To generalize: it is natural–but quite wrong–for the casual observer to believe that the two sides are equally connected to reality and equally arguing in good faith.

    I think House is going through this education as we speak.

  6. Well, GDP is a fairly good index of well-being but a country with a high GDP can be inferior to a country with lower GDP but less crime, less pollution, better health and so on. Even folks who have not read their Stiglit should see this.

    Concretely this means that Sweden is a better place to live than the US. Given that the Nordic model is far more to the left than the Democratic Party of the US (which is in fact more of a centre-right party … and if some party soldiers really wanna claim that a party which destroyed habeus corpus and grants a president the right to kill American citizens without due process is a fine progressive party I’d have to call them dumb) the facts speak for themselves: social democracy aka moderate capitalism works far better than unfettered anarcho capitalism.

    • I couldn’t agree more about GDP as a measure of well-being. We too often assume that we should be in the business of maximizing GDP which is just sloppy. If GDP entails additional costs (you mentioned crime, pollution but I’m sure there are more) then there is a trade off there that we need to respect. The Djankov/Shleifer studies are at best suggestive — not decisive.

    • “moderate capitalism works far better than unfettered anarcho capitalism”

      If you think what we have in this country is anything close to “anarcho-capitalism” you’re sorely mistaken. And I’m not saying this as a conservative, but as a lefty who would prefer something much closer to anarcho-capitalism that what we have now.

      • @Steve: I disagree. It is totally natural that anarcho-capitalism, i.e. capitalism without democracy and the rule of law, descends into some form of socialism for big corporations.

        The core theoretical problem of anarchism/libertarianism of any kind is that without a decent set of rules and a government which implements AND OBEYS those rules there is just a power vacuum … and the strongest aka richest guy will simply take it.

    • Also, we also need to be careful about comparing countries to begin with. Why are we comparing the US (a diverse nation with 300 million people) with nations in Europe anyway? Sweden has only 10 million people or so. Perhaps the comparison should be, e.g., Massachusetts and Sweden or Connecticut and Sweden?

      • I think that key difference between the US and Sweden is that the former is an immigration country whereas the latter is not. So social cohesion is higher in Sweden and higher social cohesion leads to broader public support for a welfare state.
        I don’t wanna make bold claims about an undeniable trade-off between immigration and the welfare state but I also think that this is an issue which is often ignored by liberals because of political correctness (if you say that more immigration is bad because it might lead to less cohesion and less public support for the welfare state you are quickly l labeled a racist).

      • @Keynes Still Matters
        That’s a common misconception, immigration in Sweden has been as high as in the US (or higher) for many decades now and a large share of the population are immigrants or have immigrant backgrounds.

        A far more likely explanation for the difference between the two countries is the American ‘national myth’ of liberty and rugged individualism in contrast to the consensus-seeking culture that exists in Sweden.

      • True but the Swedish welfare state has also been under attack for quite some time.

        I totally agree with you about American individualism and Swedish consensus-seeking (Ingmar Bergman has once said that the great thing about Swedish social democracy is that it makes capitalist and socialists sit together on a table and reach a compromise). But my point is precisely that these mentalities are subject to change. And less social cohesion obviously makes people become more individualistic (or tribal).

        It should be obvious that this is a descriptive and not a normative statement, I would like to live in a society where both is possible, migration and a decent welfare state. But if reality conflicts with my preferences and presents me with a trade-off I gotta make a choice. Pretending that there is no trade-off is liberal ignorance based on PC:

  7. Alternatively, it could be that the accumulation of facts may have forced liberals to adopt a more conservative tone as our understanding of the world was refined. Republicans in contrast could afford to adopt a less realistic view that was even further to the right. I thought initially that Noah’s quote was more like the latter view but I’m not sure…

    Yep. I think that in 1980, center-right policy ideas were mostly (though not all) better than center-left ones, so both the electorate and the policy consensus swung to the right. Conservative policies were implemented, mostly successfully – deregulation and tax cuts. Then we reached a more-or-less appropriate level of regulation and taxation by 1990, but conservatives, emboldened by their new majority coalition and their record of success, overreached, cutting taxes unsustainably (under Bush II) and deregulating the finance industry too much (under Clinton). In the meantime, the failure to reform our health care and immigration systems and to repair our highways became more salient, so center-left ideas became better, and both the electorate and the policy consensus started shifting to the left.

    This is my oversimplified pocket history of recent American political economy. I see it as a basically healthy and rational process, plagued by institutional momentum.

  8. Yglesias seems to cherry-pick small passages out of an econ textbook to support his conclusion about a “liberal bias,” whereas in his original post Chris picks broader points of econ 101 to arrive at his (more valid in my opinion) conclusion of a broader “conservative bias.”

    There’s a reason why liberal students felt the need to walk out of intro econ a couple of years ago at Harvard, while it would be hard to imagine conservative students doing the same (even if the professor was Krugman rather than Mankiw). I believe Yglesias has said has never taken an intro econ class, and it shows (not to say he isn’t well-read). The broad conclusion you generally walk out of introductory econ is about how markets function. While some instances of market failure may be discussed in the course, they definitely aren’t the fundamental points of the course.

    At my school, which has a notoriously notoriously liberal student body, it was clear that the liberal students in the class were having their worldview challenged more than the conservatives based on their reactions.

      • Introductory econ, sometimes called a “principles” course, is mostly about principles, which is to say, it’s mostly theory, not empirics. Many liberals would say without irony (and with some justification) that economic theory has a conservative bias. (Indeed I think many liberal economists believe there is such a bias but regard it as a sort of necessary evil, taking the view that the method of analysis is useful despite its bias and that it has no “neutral” substitute.) Whether economic facts have a liberal or conservative bias (an inherently ironic proposition either way) is quite a different question.

      • I can agree with that Andy. I was mostly just taking issue with the cherry-picking in Yglesias’s argument. I put “liberal bias” and “conservative bias” in quotes because of the irony you alluded to.

      • Yep, labo(u)r productivity is the same in both countries; however to be fair I should add the nuances that the data are now 14 years old, and that the “hours worked” criterion favourably omits France’s higher unemployment.

  9. Pingback: The facts are capitalist | Political Eh-conomy

  10. I’m not sure what to make of the doughnuts/cigarettes argument. It only makes sense to me if you think there are increasing (e.g. quadratic) costs (which there would be, for example, with taxes, and maybe with doughnuts, but I’m not so sure about cigarettes), but I still think it’s wrong. If costs are linear, then it’s perfectly reasonable to keep making the argument, “This policy has non-trivial benefits, and the costs are small enough to ignore,” with one policy after another (though one can of course question the premises in each case). If costs are, say, quadratic, then if you argue successfully for one policy on this ground, it raises the hurdle for the next. But I don’t see liberals having been successful at a long succession of incremental arguments. Rather, there is back and forth: Clinton’s tax policies raise the hurdle for future taxes; Bush’s lower it; and so on.

  11. Pingback: Baby Steps … | Orderstatistic

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s