Our new working paper, co-authored with Thierry Verdier, received an unexpected amount of attention from the blogosphere — unfortunately, most of it negative. The paper can be found here, and some of the more interesting reactions are here, here and here . A fairly balanced and insightful summary of several of the comments can be found here.
We are surprised and intrigued. This is the first time, to the best of our knowledge, that one of our papers, a theoretical one at that, has become such a hot button issue. Upon reflection, we think this says not so much about the paper but about ideology and lack of understanding by many of what economic research is — or should be — about. So this gives us an opportunity to ruminate on these matters.
First a little about the paper: There is an influential and very interesting line of work in political science (with some offshoots in economics), sometimes referred to as the “Varieties of Capitalism” literature, most often associated with the work of Peter Hall and David Soskice. The main idea is that there are many different ways of organizing a capitalist economy, perhaps with two polar cases being a Coordinated Market Economy (CME), which captures certain salient features of Scandinavian countries, and a Liberal Market Economy (LME), proxying for a US style economy. This literature suggests that both of these institutional/organizational arrangements can lead to high incomes and similar growth rates, in particular, because LMEs generate radical innovations, particularly in sectors such as software development, biotechnology and semiconductors, while CMEs are good at incremental innovation in sectors such as machine tools, consumer durables and specialized transport equipment; as part of this institutional arrangements, they also provide more social insurance and generate less inequality. The argument is that different societies have developed these arrangements for historical reasons and once established, they tend to persist (perhaps because of institutional complementarities or because of the usual difficulties of changing institutions). Because their economic outcomes are similar but CMEs provide better social insurance to their citizens, if an LME could turn itself into a CME, this would be associated with a significant gain in welfare.
The main idea of our paper is to observe that international linkages are absent from this picture, and to suggest that the institutional choice of that society is not entirely independent of the choices of others.
For one, countries trade and this induces specialization. If there are some complementarities between specialization decisions and certain institutional arrangements, the world equilibrium might be asymmetric: some countries would choose LME specialize in sectors in which this creates a comparative advantage while others choose CME and specialize in other sectors.
Another international linkage is technological, and this is the one our paper develops formally. We study a simple model in which all countries potentially contribute to advances in the world technology frontier and in turn build on this frontier in their innovation and production. We then introduce a very standard model of incentives: effort in innovative activities is forthcoming as a result of the rewards to this effort. This in particular implies that a greater gap of incomes between successful and unsuccessful entrepreneurs increases entrepreneurial effort and thus also a country’s contribution to the world technology frontier.
The main theoretical result of the paper is to show that under these assumptions and certain plausible parameter restrictions, technological linkages will also lead to an asymmetric world equilibrium in which some countries opt for LME-style institutions (what we call “cutthroat” capitalism) while others adopt CME-style institutions (what we call “cuddly” capitalism). Our model verifies the plausible notion that those under cuddly capitalism will be poorer but happier because of the greater social insurance that these institutions provide. The surprising result — which as we explain below is the main reason for writing models in the first place — is this: in equilibrium those under cutthroat capitalism cannot switch to cuddly capitalisms because of the interdependences in the world economy: when others are doing cuddly, it’s a best response to do cutthroat because the cutthroat country is contributing disproportionately to the world technology frontier and a switch from cutthroat to cuddly would slow down world growth. But given the cutthroat strategy of one (or a subset of countries) pushing the world technology frontier forward, it is a best response for others to cuddly.
So why all the furor?
We cannot help but think that some of it is a little “ideological”. In the same way that we got criticized by the right for arguing that inequality is often created by elites and that colonial exploitation has cast a long shadow on the development of much of the world (see for example this particularly outrageous rightist attack), the moment we raise some potential issues, albeit just theoretical ones, about the Scandinavian model that the left cherishes, all hell breaks loose on the left. Perhaps this is fair game.
But we think this is only part of it. Another aspect is the divide between what the academic research in economics does — or is supposed to do — and the general commentary on economics in newspapers or in the blogosphere. When one writes a blog, a newspaper column or a general commentary on economic and policy matters, this often distills well-understood and broadly-accepted notions in economics and draws its implications for a particular topic. In original academic research (especially theoretical research), the point is not so much to apply already accepted notions in a slightly different context or draw their implications for recent policy debates, but to draw new parallels between apparently disparate topics or propositions, and potentially ask new questions in a way that changes some part of an academic debate. For this reason, simplified models that lead to “counterintuitive” (read unexpected) conclusions are particularly valuable; they sometimes make both the writer and the reader think about the problem in a total of different manner (of course the qualifier “sometimes” is important here; sometimes they just fall flat on their face). And because in this type of research the objective is not to construct a model that is faithful to reality but to develop ideas in the most transparent and simplest fashion; realism is not what we often strive for (this contrasts with other types of exercises, where one builds a model for quantitative exercise in which case capturing certain salient aspects of the problem at hand becomes particularly important). Though this is the bread and butter of academic economics, it is often missed by non-economists.
Now in this light, let us turn to the main criticisms, which came in four flavors.
The first is that the assumptions were not a literal description of the world. For example, this blog picked on the fact that in the model, cuddly capitalism involves no income inequality (if you’re not going to provide strong incentives and effort is binary, then no need to create unnecessary income and consumption variability). It argued that this would be a model of socialism not of social democracy. Well this just totally misses the point of simplifying assumptions. Of course one could generalize the model to have multiple levels of effort (or more elegantly, a continuous effort choice) and then even cuddly capitalism would provide some incentives to entrepreneurs and would involve inequality between successful and unsuccessful entrepreneurs. Why didn’t we do this in the paper? Simple: this is the sort of thing that complicates the exposition and the mathematical analysis but doesn’t add all that much, so the general preference is often towards stripping these features out to distill things to their bare — and clearest — minimum.
The second criticism (for example this) questions the premise of the paper on the basis that the US wasn’t less innovative when it was more equal in the 60s and 70s and several other unequal countries are not very innovative. This is a good point, but it again takes certain aspects of the model more seriously than one should. First of all, in reality of course reward structures are not chosen optimally but result from a variety of economic and political factors. The model simplifies things by assuming that a benevolent social planner at the country level makes these choices (with the explicit caveat that this is a simplifying assumption). In reality then there will be many nations away from the “optimal inequality” beyond what would be necessary to provide incentives to entrepreneurs. As we have argued in several forums (for example here and here), US inequality has at least in part political roots and causes severe challenges to the (already dwindling) inclusivity of American institutions. So current inequality is likely way beyond what would be necessary to provide the right sort of incentives to entrepreneurs. Second and equally importantly, the mechanism in our paper clearly refers to inequality among entrepreneurs, whereas a lot of the inequality in the United States is among workers. It should be obvious that providing a safety net at the bottom of the distribution will not be a major factor in innovation decisions. Perhaps this should have been made clearer in our paper, but then again the paper wasn’t written as a general commentary on inequality but for developing a novel theoretical point.
Cross-country comparisons should be handled with care for the same reasons. But there is an additional problem in this case. As is the main point of Why Nations Fail, some nations are unequal precisely because their politics and economies are dominated — at least in part — by elites, and this type of inequality created by extractive institutions shouldn’t and doesn’t lead to innovation. Our purpose in focusing on Scandinavia and the United States was to zero in on countries that are clear examples of relatively inclusive institutions and develop some new hypotheses on why there are so much apparent differences in the inclusive institutions of Scandinavia and the United States.
The third questions the assumption that greater inequality should really lead to more innovation, for example, because better insurance might encourage entrepreneurial risk-taking. This is a fair point. Clearly, in a world in which there are no financial contracts, inequality can be too high, involving very low consumption for failing entrepreneurs, and can consequently discourage entrepreneurship altogether (or attract the wrong people and entrepreneurship). In such a world either financial contracts or other financial arrangements such as venture capital that enable better risk sharing, or if such private arrangements aren’t possible, government intervention that improves risk sharing, could encourage entrepreneurship and risk-taking. We certainly wouldn’t want to deny this point. However, the main argument here is that still some amount of reward for successful entrepreneurship is necessary for entrepreneurs to try hard and put effort for success. So our argument, again without making any judgment on US inequality being optimal by any metric (clearly it isn’t as we have argued), just agreed with what Alexis de Tocqueville noted more than 200 years ago when he wrote (see here p. 323) :
Thus inequality itself will work to forward the wealth of all, for, everybody hoping to come to share the privileges of the few, there would be a universal effort, an eagerness of all minds directed to the acquisition of well-being and wealth. Make of this nation a huge center of commerce, so that the chances of attaining the wealth with which all the rest can be obtained, multiply infinitely and ever give the poor thousand hopes, and so a thousand reasons for remaining satisfied with their lot.
The fourth (for example Matt Yglesias here as well as again this) question the data we offered in the introduction to illustrate that the United States is more innovative than Scandinavia. Yglesias is right in pointing out that patents only capture some specific types of innovation and many important breakthroughs in productivity will not show up there. But this is not an excuse for ignoring the wealth of data on patents, especially given that the economics literature has shown how patents correlate with growth both at the aggregate and firm level. Of course patenting can vary for a variety of reasons across countries. That’s why, in providing motivating evidence in the introduction of our paper, we did not just start with this next picture, which shows greater patenting activity in the United States than in Scandinavian countries.
Instead, we constructed this picture, which looks at patents by citation. In particular, it plots the numbers of patents granted per one million residents for Denmark, Finland, Norway and Sweden relative to the United States between 1980 and 1999, with each number corresponding to the relevant ratio once we restrict the sample to patents that obtain at least the number of citations (adjusted for year of grant) specified in the horizontal axis. If the difference between Scandinavian countries and the United States is driven by spurious patents or other extraneous factors leading to different patenting propensities, or if Scandinavians were engaged in innovations that were as important for the world technology frontier as US firms, then we should see the gap between Scandinavia and the United States start closing when we look at more and more highly cited patents. In fact the data show the opposite: when we look at more highly cited patents, the gap is greater. Sure this is not everything, but it does suggest that there is less of the breakthrough-type innovations coming out of Scandinavia than the United States (not that everything here is per capita, so the pattern is not driven by the fact that the United States is more populous).