Time to ditch Rawls?

The seminal work in political philosophy for the era of globalization is John Rawls’ “The law of peoples” (LoP). It was written in 1999. I was attracted to it in the early 2000s as I was then already working on global inequality which was a totally new topic, entirely ignored, in economics. The only thing that came remotely close to it in economics was the Heckschler-Ohlin-Samuelson trade theorem whereby wage inequality should go down in poor countries and up in rich countries when they engage in trade. But this was such a small part of globalizations: it dealt with wages only and left out other types of income; it left out capital flows, development aid, migration, outsourcing. None of that was discussed in economics (and mostly still is not) in a global framework as opposed to 2-country 2-goods very limiting framework. But political philosophers have thought more about it.

In the late 1990s, John Rawls turned his attention from how a single nation-state should be organized (as in his Theory of Justice, ToJ) to how the world should be organized. Obviously, it had to be done at a very abstract level and, as will be seen, that abstract level comes very close to the situation Rawls (and the world) seemed to face in the 1990s. But, and it will be my key point here, that situation has dramatically changed in the past 20 years so that the abstract sketch of the world made by Rawls is no longer  compatible with what we see today and thus the recommendations Rawls drew from that sketch are irrelevant.

In LoP, Rawls abandoned the metaphor of individuals meeting behind the veil of ignorance to agree on an a priori basis on the principles of justice in their societies. That rule still holds, according to Rawls, in individual societies but not in the world of nation-states whose representatives (but not individuals themselves) meet to agree on the principles that would guide their (inter-societal) relations. 

Rawls has five types of societies:  liberal (these are the same societies with which his ToJ is concerned), consultative hierarchical societies, “burdened” societies, outlaw states (notice: not societies) and benevolent absolutisms. We can drop the last because they never play a role in LoP (I never understood why; perhaps Rawls just did not know what to do with them). Both liberal and consultative hierarchical societies are well-ordered societies (meaning that within each of them the principle upon which they are based are reinforced by peoples’ daily actions); they respect each other and the (different) principles upon which each is based. Burdened societies cannot become liberal because they are held back by their poverty. Outlaw states just go to war (for basically no reason; they just, like in a Hollywood movie, seem to like being troublemakers).  

So the rules then become relatively simple—and some would even say simplistic. The well-ordered societies, though different in their internal structure, can coexist at peace because they respect each other, and liberal societies do not try to impose their norms on consultative hierarchies. They do not try to export democracy. 

Second, liberal societies have a duty to help the burdened societies but only as far as they require to become liberal, which, in Rawls’ view, takes place at a very low average level of income. After that point, even huge differences in incomes within the group of well-ordered societies are no ground for continuation of international aid. In other words, there is no reason for Norway to help Bangladesh because they are both well-ordered. 

Finally, Rawls is against migration as a right or against migration as a way to alleviate global poverty and inequality. Countries (that is, organized peoples) have control of their territory and they alone decide whom they want to accept. They can accept refuges that flee persecution but not economic migrants (which is by the way consistent with Rawls’ general underplaying of the importance of income for our happiness).

So, this is the sketch: liberal societies reaffirm their liberal principles daily, they live in peace with hierarchical societies, the do not export democracy, they help only the poorest countries and this very moderately, and they do not allow economic migration. 

As you can see, this is why I was attracted to Rawls: unlike economists he does present a coherent sketch of the world and economic rules.

So why do I have a problem with Rawls’ taxonomy now?

Let me list several changes that have happened during the past two decades and for which I just do not find a place in Rawls’ taxonomy.

Liberal democracies do not affirm the principles of liberalism, as Rawls expected, neither domestically nor internationally. It was inconceivable for Rawls, when these societies are working well, that they would, as in the US now, generate a third or more of “malcontent” population that clearly does not believe in liberal principles nor is willing to affirm them in their daily lives. Far from it. This, plus the pervasive role of money in electoral politics, lower tax rates for capital than labor, neglect of public education etc. imply that domestically so called liberal societies are very far from Rawls’ idea of liberalism. The difference is so great that we cannot, I think, speak of the discrepancy any longer as the expected difference between an abstract idea and what exist in reality. These societies belong to an entirely different category.

Moreover, in foreign policy, as became clear with the Iraq war, they act like outlaw states since they break the fundamental rules on which the international community is founded, namely absence of wars of aggression.

Thus, “liberal” societies are both non-liberal (in the Rawlsian sense) domestically and act as outlaw states.

The benign consultative hierarchies that Rawls had in mind probably in order to fit Islamic societies in his scheme are practically non-existent.  The Middle East is either in total chaos or in the grip of absolutist dictatorships like Egypt, Saudi Arabia and the Gulf sheikdoms.  Thus they are not well-ordered societies in Rawls’ terms.

There is no place in his taxonomy for the multi-county non-state organizations like ISIS. A general theory that has no place for organizations that do not accept current state borders is clearly incomplete. (This is an issue on which Rawls is especially weak because he takes borders as given, which is, as he wrote in the wake of the break-up of the USSR, Czechoslovakia, Ethiopia and Yugoslavia, rather odd).

There is also no place for what is now called illiberal democracy, namely a society that has most of the accoutrements of a liberal society (elections, political parties, NGOs), and yet where only one party or one leader ever wins elections and where the media and the judiciary are directly or indirectly controlled.

Migrations, driven by economic reasons and thus by global inequality, do not have a place in Rawls. But they do exist in real life where economic migrants from Africa and Asia into Europe or Mexico and Central America into the United States number millions. But the theory that says that this should not happen is of no use when these things do happen.

Finally, Rawls grossly underrated the importance that people attach to income and wealth for their happiness. Importance of pecuniary incentives has only increased with globalization since income differences have become more visible.    

The changes over the past two decades have been, I believe, so remarkable that the typology offered by Rawls has lost its relevance. But if the typology does not fit reality then the recommended relations between the different societies, based on this typology, do not have any relevance. This is why I think it is time to either ditch Rawls or revise him very thoroughly. A job for political philosophers.

The forthcoming changes in capitalism?

Sometimes it’s useful to put symbolic dates on when a different era begins. The end of Thatcherism, it could be argued, came on July 10 in the then PM-candidate speech by Theresa May. It was perhaps appropriate that another woman, a Tory Prime Minister, would be credited with the ending of Thatcherism. The key words, which immediately attracted attention (see also Philip Stevens in today’s “Financial Times”) were not those about  inequality (which has become a common place these  days) but about the changes in the internal structure of capitalism: reintroduction of workers’ and consumers’ representatives on management boards, limits on the executive pay, reduction of job insecurity for the young people and much greater access to top jobs for those coming from less privileged backgrounds.

For the first time since the late 1970s (at the top level of policy-making), we are back to the issues of reforms in the way capitalism functions rather than discussing the ways in which the external environment would be made more market friendly. In essence, this is a confession that “civilizing” capitalism cannot be done only “externally” by relying on the “harmony of private interests” but that the state has a bigger role that goes beyond ensuring the protection of property rights, taxation and redistribution.

The past 35 years have shown that the neo-liberal conception of capitalism, combined with its global reach, has increased inequality to often unsustainable levels, left large segments of the population in the rich world without significant increase in real income and with heightened insecurity, and brought populist policies with a vengeance. Theresa May was thus right to highlight that the Brexit vote was at least as much a vote about the British economic system and the role of the political elite as it was about Brussels.

It may be too early to see exactly how things may change but I can imagine the main changes in three areas.

The first area is economic policy of the advanced countries. The key disagreement there is between those who believe that the current economic dissatisfaction and the populist backlash are the product of the 2007-08 crisis and those who call for more fundamental changes.  The first group believes that the problems will vanish once Western economies go back to growing at 2-3 percent per year and reduce unemployment (as the US has already done). I think they minimize what has happened in the meantime and  lack any ideas how to address the structural weaknesses of capitalism. This is where Theresa May steps in by arguing for structural changes (the structure is meant here as the way capitalism works).

I have argued in my “Global Inequality” that taxation of current income seems to have reached the maximum that the rich countries’ electorates can tolerate. Therefore, reduction of inequality, “empowerment of ordinary people” and widening of opportunities have to be accomplished through a change in the distribution of both financial and human assets. This implies deconcentration of capital ownership (which in all rich countries attains almost unfathomable Gini levels of 0.85-0.9) and much fairer access to high-quality levels of education (and thus to high pay) to those born in poorer families.

On the first point, it is often said that the middle class has no interest to acquire capital mostly because of volatility in its returns. But if as much ingenuity had been deployed in finding fiscal ways to make diversification of ownership appealing to more people as it was invested in making the rich create tax  loopholes, one would be able, in the medium term, to reduce the current extreme concentration of wealth.

The second area where I see the need for major changes is in development. It has always been the case that the development economics (including development economics as applied by major international organization) has moved together with economic ideology waves in rich countries. In full ideological lock-step with neo-liberalism, the past several decades in development were dominated by efforts to “improve the economic framework” (make regulation lighter, allow faster registration of enterprises, reduce government red tape etc.)

While many of these policies made sense they had two significant shortcomings: they exaggerated (again fully in agreement with similar policies pursued in advanced economies) the ability of the free market to spontaneously produce desirable outcomes, and they shifted the role of international agencies toward the “soft” areas, from  gender and sexual-orientation equality to “transparency”,  to the detriment of actual investments in roads, infrastructure, telecommunication, electricity generation. Indeed, the “framework goods” are important, but they are not all that matters. Moreover, this emphasis led to millions of hours of (wasted) work and thousands of papers that just repeated nostrums, creating NGOs they preyed on foreign and their citizens’ money to pay themselves hefty fees under the pretense of doing the work of development. Such lop-sided emphasis has made organizations like the World Bank consistently less important players in the area of development. The more robust Chinese approach, reflected in its policy in Africa, and now in the new Asian Infrastructure Investment Bank has given, one would hope, a good jolt to the old development institutions  to move back to the harder areas of development, and not solely on organizing conferences with “stakeholders”.

The third change that is coming regards immigration policies. Here Europe is the primary battlefield because the issue of migration has been conflated with the issue of terrorism. This is an unprecedented development. Whether we look at the European migrations in the 19^th or 20^th century to the Unites States and Latin America, Indian labor moving to the West Indies, or the Chinese moving to East Asia, migration problems were seen as those of assimilation, jobs and wages for the native labor, but never as issues of personal security. (A possible, small exception to that is the view, briefly held in the United States, that South-European immigrants at the close of the 19^th century were disproportionally likely to be anarchists.)

Major adjustments in European countries to accept more of non-European migrants are already taking place. Even if Angela Merkel cannot go back to what was already done and expel the migrants Germany accepted last year, it is clear that no such wave will be willingly accepted In Europe any time soon. (It does not guarantee that another such wave may not materialize though—but it will not be welcomed.)

But creating a fortress Europe is difficult and runs counter the economic interests of an aging Europe that needs fresh labor from Africa and Asia. This is why I have argued in “Global inequality” for a change in the approach to citizenship which is currently seen as an almost inevitable end-point for any individual who makes it to a EU country (or the United States and Canada). Instead of that, different levels of residency, including temporary jobs (that were originally the idea behind the German Gastarbeiter system) with the obligation to return to the country of origin, should become generalized. Citizenship rights should be divorced from the right to work in a country.  

If this line of thinking is correct, we should expect in the next decade a major redefinition of how capitalism works in the advanced societies, a redefinition of the role of international development organizations, and a redefinition of what it means to be a migrant and a citizen. This should, in principle, allow the rich countries to continue their key role in globalization and to be rid of this paradoxical position, where they, the spiritus movens of globalization, seem now most afraid of it.

In defense of equality (without welfare economics)

When I taught recently at the Summer School at Groningen University, I began my lecture on the measurement of inequality by distinguishing between the Italian and English schools as they were defined in 1921 by Corrado Gini:

“The methods of Italian writers…are not…comparable to his [Dalton’s] own, inasmuch as their purpose is to estimate, not the inequality of economic welfare, but the inequality of incomes and wealth, independently of all hypotheses as to the functional relations between these quantities and economic welfare or as to the additive character of the economic welfare of individuals”. (Corrado Gini, 

Measurement of Inequality of Incomes, Economic Journal, March 1921.)

I put myself squarely in the camp of the “Italians”. Measurement of income inequality is like measurement of any natural or social phenomenon. We measure inequality as we measure temperature or height of people. The English (or welfarist) school believes that the measure of income inequality is only a proxy for a measure of a more fundamental phenomenon: inequality in welfare. The ultimate variable, according to them, that we want  to estimate is welfare (or even happiness) and how it is distributed. Income provides only an empirically feasible short-cut to it.

I would have been sympathetic to that approach if I knew how individual utility can be measured. There is, I believe, no way to compare utilities of different persons. We all agree that the marginal utility must be diminishing in income because it is the foundation of economic micro theory.  (If marginal utility of income were not decreasing, we would not be able to explain why demand curves are sloping downward.) But we have no idea whether, whole both your and my marginal utility functions are decreasing, my level of utility may, at any point, be orders of magnitude greater than yours or the reverse.

The only way for the “welfaristas” to solve this conundrum is to assume that all individuals  have the same utility function. This is such an unrealistically bold assumption that I think nobody would really care to defend it except as here where it is considered as a lesser evil that allows “welfaristas” to cling to their Utilitarianism, to define measures like the Atkinson index and to continue believing that the real thing we want and (they clam) do measure is inequality in individual welfares.

Now, the welfarst approach continues to be associated with pro-equality policies. 

Why? Because if all people have the same utility function, then the optimal distribution of income is such that everybody has the same income. If from that equilibrium you take some income from A and give it to B, loss of utility of A will outweigh the utility gain of B (because marginal utility is decreasing) and thus obviously total utility will be less in any situation where income is not fully equally distributed.

My students then asked how I can justify concern with inequality if I reject the welfarist view which is the main ideological vehicle through which equality of outcomes is being justified. (A non-utilitarian, contractarian alternative is provided by Rawls. Yet another alternative, based on equal capabilities—a close cousin to equality of opportunity (of which more below) is provided by Amartya Sen.)

My answer was that I justify concern with income inequality on three grounds.

The first ground is instrumental: the effect on economic growth. After the period of the 1990s where, due to lack of data, we ended up with inconclusive results on the link between inequality and economic growth, we are  having more and more evidence that high levels of inequality slow down the increase in total income. We are able to show that now because we have access to micro data and a much more sophisticated view of both inequality and growth. Here, as examples, are the papers by Sarah Voitchovsky and by Roy van der Weide and myself. But, it has to be acknowledged: if empirical literature were to come to a different conclusion, namely that inequality helps growth we should have to drop that instrumental argument against high inequality.

The second is political effect. In societies where economic and political  spheres are not separated by the Chinese wall (and all existing societies are such), inequality in economic power seeps and ultimately invades and conquers the political sphere. Instead of one-person one-vote democracy we get one-dollar one-vote plutocracy. This outcome appears inevitable, especially in modern societies where running political campaigns is extremely expensive. But it was not very different in ancient Greece or Rome. if we hold that democracy, a more or less equal influence of everyone on public affairs, is a good thing, we have to be in favor of severe limits on income and wealth inequality. It seems to me that the negative impact of inequality on democracy, not only obvious only in theory, is now being confirmed empirically as well (see Martin Giddens’ “Affluence and Influence”).

Let me note parenthetically that even if we failed to detect such explicit influence of the rich on policy making, the a priori case that it must exist (but is difficult to measure) would still be extremely strong. Because for the opposite case to hold, i.e., rich people give money to politicians but do not get anything in return, we have to assume an entirely irrational behavior of the rich: they throw money away for no reason. That goes so much against the fundamentals of economics that if we assume it here we should also assume it that when people (say) go to restaurants; they throw money randomly: “Your glass of wine costs $10. Well, I will give you $15 and you do not need to give me the wine”. If that behavior seems reasonable to you, then I would agree that the rich may not influence politicians to whom they give money.

The third ground is philosophical. As Rawls has argued, every departure from unequal distribution of resources has to be defended by an appeal to a higher principle. Because we are all equal individuals (whether as declared by the Universal Charter of Human Rights or by God), we should all have an approximately equal opportunity to develop our skills and to lead a “good (and pleasant) life”. Because inequality of income almost directly translates into inequality of opportunities, it also directly negates that fundamental equality of all humans.  This is I think pretty evident on an a priori basis, but we have also an increasing number of papers that show the positive correlation between inequality of income and inequality of opportunity (see Marrero and Rodriguez, Miles Corak). Families with greater income ensure that their children have much better opportunities (which negates the fundamental equality of which we spoke) and in turn make sure they this new inequality of opportunity is converted  into yet higher income for themselves and their own children. So, a positive feedback works very strongly to maintain unequal access to opportunities.

I have to say here that in addition inequality of opportunity affects negatively economic growth (so we now have a negative effect going from my third ground back to the first) which makes inequality of opportunity abhorrent on two grounds: (1) it negates fundamental human equality, and (2) it lowers the pace of material improvements for society.

My argument, if I need to reiterate it, is: you can reject welfarism, hold that inter-personal comparison of utility is impossible, and still feel very strongly that economic outcomes should be made more equal—that inequality should be limited so that it does not strongly affect opportunities, so that it does not slow growth and so that it does not undermine democracy. Isn’t that enough?