27 December 2005

Economics and the Public Purpose

Review of Economics and the Public Purpose by John Kenneth Galbraith

I realize that the title seems to suggest that reading the U.S. tax code would be more interesting, but Economics and the Public Purpose is actually a great book. The writing is wonderfully crisp and engaging, and the ideas are something that everyone needs to become familiar with, even if they sometimes overreach a bit.

Galbraith is a famous liberal economist, born in Canada in 1908 and somehow still alive. He wrote this book as part of a series which also includes The New Industrial State and The Affluent Society. I haven't read anything else by him, but if I do, I'll blog about it, and perhaps one of you can tell me about his other writings? Economics and the Public Purpose was written in 1970, and arguably the corporate climate has changed since then in ways that invalidate some of the book's arguments; however, I think most are still applicable.

As the title implies, the book is framed as a critique of economics (or as he calls it, "neoclassical economics") for not serving the public purpose. In particular, neoclassical economics assumes various things about markets that (a) are false and (b) obscure and thereby naturalize and strengthen existing power structures. These assumptions survive because of (b) and also because of their conceptual simplicity. Here are a few:

ideas from neoclassical economics
  1. Firms always try to maximize profits. (It might seem paradoxical that profits are zero in perfectly competitive markets, but this isn't a big problem either in theory or practice.)
  2. Sovereignty rests with consumers, whose tastes/needs dictate demand curves, and nature, which dictates supply curves. In particular, firms have no sovereignty, since if they try to do anything other than maximize profits they'll be replaced. (Even a firm in a monopoly position will be forced by its shareholders to maximize profits. Its shareholders have to, because they're mutual fund managers who will be fired if they don't. Or because if not, some aggressive new manager could borrow a few billion dollars, perform a hostile takeover and make higher profits. You get the idea...)
  3. Similarly for government. All government ultimately responds to voter will. Or maybe voter plus lobbyist will, and maybe voters are dumb, but still government officials have no meaningful agency of their own to exercise.
I've of course phrased these in ways where it should be obvious what's wrong with them.
problems with neoclassical economics
  1. Agency problems mean that corporations/governments are run in practice by managers (which Galbraith calls the 'technostructure') that are distinct from owners/voters. Monitoring is imperfect and costly.

    This idea has been enormously productive in economics, political economy and elsewhere. For example, perfect credit markets make persistent poverty hard to explain, since the higher marginal product of capital in India (or wherever) should cause American investors to prefer it to loser GM stock. This would mean that a country's starting amount of cash should dictate only its level of consumption, while production moves to wherever it's most efficient (assuming free trade). The problem is that prospective borrowers don't have collateral, and it's too hard to keep them from disappearing with the money they've borrowed. As a result, everyone invests (inefficiently) in rich countries and poor countries stay undeveloped. Microcredit seems like a good, if limited, answer to this problem.

    There are actually a lot of other interesting stories of agency problems, but I'll save them for another post, so I can focus on what Galbraith is talking about.

  2. Firms no longer try to maximize profits. Instead their first goal is growth, as long as profits are high enough to avoid bankruptcy or a takeover. This is because profits (mostly) go to owners, but the technostructure mostly benefits from growth, through promotions, increased market power and better job security. (Arguably corporate shakeups in the last few decades have weakened this argument, though much of it still rings true.) The only time profits become an issue are when there's the possibility of takeover, shareholder revolt, or some other kind of external threat, which is not too often if things are going decently.

    Some people have a hard time believing this point. If so, here's a useful thought experiment. Would you rather run a small dry cleaning business that gets 20% returns on capital, or be the CEO of GM in a year where the company loses 1% of its value?

  3. Firms no longer react passively to consumer demand, government regulation and market prices. Instead they can shape these with advertising, lobbying/regulatory capture and by using their market muscle to dictate prices. Advertising and lobbying are obvious. Regulatory capture is when the corporate technostructure links up with the government technostructure and helps shape government action; the most famous example is the military-industrial complex, but the same principle applies to the FCC, EPA, FDA, etc...: Congress can pass laws, but the implementation has to be left to bureaucrats who can never be perfectly monitored and held accountable. Finally, having a large market share (the result of a focus on growth) means that large corporations have a good deal of freedom to negotiate their own prices. However, while a neoclassical monopoly or oligopoly should charge higher prices (and have lower sales, but higher profits), we actually see lower prices (and higher sales) since firms use their market power to promote growth rather than profits.
So (neoclassical) economics gets it all wrong. So what? What's so bad about implicit rule by the technostructure? (I should point out that Galbraith often reads like Chomsky. He has the same sweeping and blistering critiques of ideology and orthodoxy, and is refreshing in many of the same ways. Both are good remedies to Thomas Friedman, for example. Of course, he's frustrating in some of the same ways too; sometimes he treats the reader as though we've never read anything other than the standard party line, c.f. Goldstein's book. One of their main differences is that Galbraith is a liberal and Chomsky is a radical, so that Galbraith proposes solutions that, though counter-cultural, are more technical than revolutionary.)
Why rule by technostructure is bad
  1. Inefficiency: Like in USSR-style state capitalism, prices and levels of production are set arbitrarily, and therefore inefficiently. If we presuppose that a free market will maximize total welfare (first law of welfare economics), then this is in general suboptimal. However, Galbraith gives more specific and interesting problems.
  2. Underdevelopment of the market sector: First I should explain that Galbraith refers to large corporations as the "planning sector," meaning that they can exert control over prices, consumer demand, gov. regulation, etc., as opposed to the "market sector," which consists of small firms that don't have this power. Many economic activities naturally fall within the market sector and resist organization into large corporations: personal services, local businesses, artists that can't deliver standardized products, etc..

    Since the planning sector is stronger than the market sector, they get to treat the market sector like a poor stepchild, for example passing price increases on to it as they see fit. Also, the market sector is vulnerable to inflation and interest rate fluctuations in ways that the planning sector is not, as large corporations are often able to finance expansion using profits rather than debt.

    So we have less art, medicine and child care than we should, though these arguments always seem a little dicey to me. More compelling is...

  3. Overdevelopment of the planning sector: We overconsume things produced by the planning sector, like cars and Coke.

    It seems like this can't totally explain problems like suburban sprawl, though. Is it the planning sector's fault that we have too many cars and too few trains? Well, sort of, in that one part of the planning sector (car manufacturers) muscled out another part (trolley manufacturers, or whatever). But it's not like some general tilt of the playing field away from the planning sector and towards the market sector would help this.

    More interesting is that we overconsume period. Or rather, we overconsume products and underconsume services and leisure (i.e. work too hard). This is relevant to the planning/market distinction because many services naturally fit into the market sector (because small/local businesses are involved), while manufactured products tend to come from the planning sector. Advertising is one mechanism that makes this possible. Advertising has many different effects: encouraging consumption of a particular brand (Saab), encouraging consumption of that class of products (cars) and encouraging consumption in general as a solution to problems (angst, need to express personality but not knowing how). (Or I could mention the rush credo for pre-frosh weekend: "rush MIT, then rush Greek, then rush AEPi," with "rush college" left implicit.) A car company without much market share can only take advantage of building brand awareness, while a large company also benefits from new drivers entering the market, since they'll get a decent fraction of them. Thus, we expect the planning sector to advertise more heavily than the market sector. Not only is this part of their advantage over the market sector, it also encourages consumption of products in general as a road to happiness.

    However, advertising is only the crudest way that the planning sector shapes public thought. Beyond telling us what to consume, it also tells us what to think.

  4. Shaping ideology: This last point is tricky, because large corporations of course don't have Thought Police (for the most part) and outside of their marketing departments, don't usually try too hard to shape public thought. However, ideology follows power, and so it's inevitable that our values will be shaped by the planning sector: the difference between serious/frivolous, respectable/eccentric, etc...

    One example is the different way that we view science/technology and art. Advanced technology is naturally suited to the planning sector, because it relies on standardization, mass production, specialized labor, and so on, while art is not, since it's usually better if it's individually produced. Before the Industrial Revolution, art and science were considered comparably valuable, and as science and engineering became more useful to people in power, social values changed accordingly. On the other hand, maybe people just respect money, and that's why executives have higher status than performance artists. But that can't fully explain why we think some jobs should be higher paid than others; it's considered natural for artists to be poor, and in fact there's often the suggestion that their art is better if they don't expect to be paid for it (i.e. they don't "sell out"). No one would ever suggest doing the same for scientists, even if most scientists are similarly motivated more by interesting work than by money.

The way that the planning sector shapes ideology is worth dwelling on, since it's the first obstacle to reform. Galbraith uses the term convenient social virtue to describe values of the planning sector that have been internalized by mainstream culture in ways that make things cheaper, easier or more profitable (hence convenient) for the planning sector. For example, the military needs to convince millions of people to enlist and get salaries much smaller than civilian contractors doing similar jobs. This can be done on the cheap by promoting the virtues of patriotism and serving one's country; equivalently, one might say that soldiers are compensated partly by their salary, partly by their social role, which lets them be proud of themselves and gives them respect from the rest of society. For example, Vietnam vets complain that they were cheated out of the post-war respect that they considered their due. And compare the reactions of soldiers and of corporate lawyers when the work they do is criticized: soldiers need the cultural compensation in a way that people with higher pay and better working conditions never would.

Patriotism also has the side benefit of helping the government convince the population to go along with its policies, especially wars. Iraq is a good example, but in general, the political system in the U.S. finds wars almost irresistible. Dwelling on this point turns the idea of government responding to voter preference (e.g. the median voter theorem) on its head, in the same way that Galbraith critiques consumer sovereignty. It's far from a new idea (recall the Goering quote: "Of course the people don't want war. But after all, it's the leaders of the country who determine the policy,..."), but it's interesting how Galbraith unifies his critiques of modern capitalism and of modern democracy. Similarly, the Nazi political theorist Carl Schmitt famously said that "Sovereign is he who decides on the exception"” (e.g. who declares a state of emergency).

There are many other examples of the convenient social virtue, and how it enables the system to be run more cheaply. Teachers, for example, could be underpaid initially because other jobs were closed to educated women, and later by representing teaching as a form of national service, as in Teach for America. Similarly for social work. (To see this, observe that if these jobs had competitive salaries, we wouldn't need to think of performing them as "service.")

Another of Galbraith's examples is the idea of the hard-working small business owner. He says that while those with comfortable jobs in the planning sector wouldn't accept unpredictable unpaid overtime, this is a natural part of the life of a small business owner. They put up with it because of the status afforded to entrepreneurs, rather than for the financial benefit. This one seems kind of dubious to me, mostly because I've forgotten his argument as to why it should help the planning sector, but also because even highly-paid people in the planning sector (like consultants) often work long hours, and because there are natural reasons for small businesses to be more flexible than large ones in many things, including demands on their workers.

Finally, he describes patriarchal consumerist family life ("the American way of life") as a convenient social virtue that's key to most of the others. The idea is that maintaining a high level of consumption requires women to stay at home to organize it all (an assertion which I don't think has aged well), but also that suburban family life encourages consumption through competitive pressure. It's seen as virtuous for women to take care of kids and do housework (w/o much pay) and for men to work hard to support their families. (An alternative choice might be for both parents to work part-time, reduce their consumption of goods and raise their consumption of services, including outsourcing housework. Or the traditional family might be rejected altogether.) Neoclassical economics overlooks all of these issues by making the "household" the unit of analysis rather than the individual.

The problem with this whole "convenient social virtue" discussion is that agency often gets confused, and it oversimplifies to say that the planning sector both benefits from and creates these convenient social virtues. The examples Galbraith gives are good starting points, but a Foucault-style critique is probably more appropriate. For example, The Wages of Whiteness is one long examination of how White supremacy became dominant in 19th century America; White pride is mainly considered a convenient social virtue for lower-class White workers (i.e. they receive social/psychological "wages" from their Whiteness), but the book goes on to say some nonobvious things about the origins of this racism. If/when I blog about it, I'll explain in more detail.

This post is getting long, so I'll skip to the punchline.

how to make things better
He starts with a section called "The Emancipation of Belief," which says that we need to actively resist advertising and propaganda that supports the values that come from the planning sector, instead of imagining that we're protected by cynicism about the more outrageous claims of advertising. Of course, this isn't really a personal project, and he's a little vague about how to pursue it with most of civil society in the hands of the planning sector, but it's a good start. And he also says that universities are a good place to organize around, since the economic necessity of critical thinking in universities will preclude any 1984-style repression.

The next step is to use the state (specifically the legislature) to restore economic parity between the planning and market sectors. This means easy credit from the central bank (since the planning sector finances expansion with cash and only the market sector needs credit), precisely targeted price and wage controls (since the planning sector is already controlling prices and wages), a universal living wage (he responds to the claim that it'll encourage unemployment by saying that unemployment is preferable to degrading low-wage work once you've rejected the convenient social virtue which says otherwise), and various other liberal reforms. In an era of New Democrats, New Labour (with new New Deals), etc., it's refreshing to hear such an unapologetic and compelling defense of big-government liberalism, even if it's not all completely convincing.

Most of his arguments are still relevant in one form of another, and the writing is infinitely better and more enjoyable than the brief summary I've given above. Read it!

15 December 2005

Les causes des émeutes en France

For my evening French class I wrote a short piece about the riots in France. Since it's as interesting as anything else I post here, and since pretty much all my writing ends up either here or on the arxiv, I thought I'd share it. Plus, this way if I said anything stupid about France, someone here can correct me.

read French version of essay

Commençons avec les problèmes economique des immigrés et leurs descendants. C'est commun dans tout le monde que les immigrés arrivent pauvres, et soubirent la discrimination d'emploi et d'enseignement. En suite, c'est plus sur que leurs enfants manqueraient aussi de l'argent et de l'enseignement. Peut-être a cause des garanties d'emploi en France ce sera plus difficile de trouver l'emploi pour ceux qui ne sont pas déjà bien branchés. Et peut-être le fait que le systême d'enseignement français demande a une jeune âge des decisions importantes donne une avantage aux enfants avec des parents instruits? D'un autre coté, France a une système de protection sociale mieux que la plupart du reste de la monde. Mais peut-être ces programmes ne suffisent pas pour des communautés entières.

C'est improbable que tous les causes des émeutes sonts economiques. Le racisme est commun partout, mais en France elle est mélangée avec un nationalisme dangereux. Les vues racistes sur les français descendus des arabes ou des africains n'est pas qu'ils sont inférieurs, mais qu'ils ne sont pas vraiment francais. Ces vues sont associées avec la croyance du 19ème siècle qu'il y a une vraie identité française qui est culturelle et raciale, est qu'il faut la protéger des menaces étrangères. La forme la plus dégeulasse vient de la Front Nationale, dont les partisans disent que les immigrés (généralement non-Européens) constituent ce menace. Mais les formes de nationalisme defensifs plus bénins essaient aussi à protéger les fromages français contre les réglementations Européens, la culture populaire française contre l'Hollywood et la langue française contre l'empruntage des mots étrangères.

D'un autre coté, le nationalisme française essaie à éffacer les differences a l'intérieur de la France. Il n'y a pas de statistiques raciales, des lois contre la discrimination raciale, et surtout il n'y a pas une compréhension de l'identité comme quelque chose fondamentalement multiculturelle. Un exemple célèbre est la phrase ``nos ancêtres, les Gaulois,'' qui apparaissait dans les livres scholaires aussi tard que les années cinquantes dans la France et dans ses colonies. Malgré le fait que la plupart des français blancs ne sont pas descendus des Gaullois (leurs ancêtres etaient plutôt allemands), cette mythe culturel dit que les citoyens français avec des ancêtres africains (ou arabes, or portugais) ne sont pas vraiment français.

Cette contradiction en l'identité française est une raison que les émeutes sont si inquiétantes; elles sont un signe qu'il faudra laisser tomber les idées de la purété raciale et culturelle pour que on ait une société inclusive et democratique.

see my notes (in English) for the paper

for example, many immigrants arrive poor, face job + education disc. and pass on poverty and lack of education to their children. perhaps job security in france makes this worse for those without jobs? and perhaps the fact that the education system forces early decisions may also give an advantage to children with educated parents, and may make it harder to escape poverty, especially for culturally marginalized groups. on the other hand, france has better social programs than many countries; [remainder changed to ``but maybe these programs aren't enough to help entire communities.''] however, it is possible that they merely ameliorate the worst effects of poverty w/o helping people rise out of it. the problem of unemployment is not just a lack of income, it's also a lack of social integration.

codeterminted, but probably more significant is the problem of racism, here blended with french nationalism. racism directed against people of arab/african descent is not so much that they are inferior, but that they are not ``french.'' correspondingly, there is a strong notion of french identity that is reminscent of the 19th century ideas about racialized nationalism. on the one hand, we see frequently the idea that french culture, language, society and identity are under threat from the outside. the ugliest form of this is from the national front, who argue that immigrants constitute this sort of threat, and today call for their deportation. but milder forms of defensive nationalism also try to protect french cheese from EU regulations, french popular culture against Hollywood, french language against borrowing words from other languages.

on the other hand, french nationalism tries to erase differences within its borders. no statistics of race, no laws against discrimination, and most of all, no understanding of french identity as fundamentally multicultural. even in the 19th c, racial purity was a myth, and as late as the 1950's textbooks (in both france and her colonies) referred to ``our ancestors, the Gauls.'' (and even white French mostly have German ancestry - "nos ancêtres, les Gaullois" and general sentimentalizing of the Gauls were only invented ca. 1789 b/c they fit the politics of the time.) but these myths are the reason that french citizens whose ancestors are from africa (and not Gaul, poland, or portugal) are not considered french.

this refusal to acknowledge race is also why the riots are so troubling; they are a sign that ideas about racial/cultural purity will need to be dropped in order to establish a democratic and inclusive

Earlier I also prepared a talk for the class, which summarized a talk I had heard a year earlier about first & second-generation female Arab immigrants in France. Here are my semi-grammatical notes (in French), but you should really just google the prof I heard the original from to read more.

Pour mon discours, je vais resumer une conference que j'ai attendu l'année dernier. La conference s'appellé "des beurettes -- aux descendantes d'immigrantes nord-africains" et c'était donnée par Nacira Guenif-Souilamas, qui est une sociologiste a l'université de Paris.

Le mot "beurre" est verlan - on l'obtien par renverser les syllabes du mot "arabe" - et "beurette" veux dire une femme arabe. Déjá en regardant les mot, on peux voir que, quand on parle des jeunes arabes, souvent on s'occupe seulement avec les jeunes hommes arabes. On voix ca aussi avec la discussion sur les émeutes maintenant. Mais les jeune femmes francaise-arabe ont leurs propres perspectives, et leurs propres problemes.

La lecture que j'ai attendu était fondé sur des centaines d'entretiens que prof. souilamas a fait avec des jeunes femmes qui étaient nées en france avec des parentes nord-africains. Je vais concentrer sur seulement une de ses arguments, qui est que les voix dominantes en france disent qu'il faut proteger les femmes arabes contre la sexisme de la culture arabe et musulman. Souilamas a dit ensuite que cette position n'aide pas beaucoup les jeunes femmes arabes. Premierement, etre émancipé de la culture arabe est souvent quelque chose que elles ne veulent pas. Souvent elles choisissent les hijabs (head-scarves?) que leurs meres ont abandonné, ou elles sont plus conservative sexuellement; peut-etre comme facon de s'exprimer et revendiquer leur identité (well, to express themselves by asserting their culture..). Deuxiemement, l'injoction de devenir francais enseigne aux jeunes femmes que leur culture ne vaux pas beaucoup; donc quand elles écoutent cette message, elles parfois acceptent et intériorisent cette racisme. Au meme temps, ca ne le rends pas plus facile assimilation, parce-que il reste encore la discrimation. Troisiemement, la sexisme qui vient des hommes arabes n'est pas souvent la problem la plus pire qu'elles ont. Souvent les hommes faisent pire en école et ont plus de chomage, et puis n'ont pas beaucoup de pouvoir pour maitriser les femmes. L'état, au contraire, peux enlever des droits plus facilement, comme la loi recu qui a interdit les hijabs dans les écoles. On n'osera pas interdire des practiques d'hommes, mais pour "proteger" les femmes, on enleves leur autonomie.

Tout ca n'est pas pour dire que c'est toujour mieux pour les jeune femmes arabe en france de choisir la culture arabe au lieu de la culture occidentale. Mais, ca sera idéale si elles avaient une vrai choix, et si elles seraient respectées n'importe quel choix elles ont fait.

02 December 2005

Experimental validation!

Publications are all well and good, but you only really know that you've arrived in physics when you (and collaborators) propose some "arbitrarily accurate composite pulse scheme" and some respected experimentalists actually implement it! "Hot damn!" you say? And rightfully so, but let's see their conclusions.
The more complex B4 and P4 sequences, although theoretically superior, do not perform well in practice.
Oh well. It can be hard not to take this personally, and feel, after long days in front of the blackboard/web browser, that as a researcher and even a person, I am "in practice quite poor," "less useful than initially expected" and even "highly sensitive to the presence of off-resonance and phase errors." Happily, though, I'm still "theoretically superior"!

On the other hand, I did recently propose a "bubble-collapse" theory that explained the noise electric kettles made, and after some pointless arguing about rival theories, designed and carried out an experiment that proved I was right. (Stir it and the noise goes away!) Nevertheless, it's probably good that I've moved to CS.

Beware the Phase Errors, my son!
The sigma_X that bites, the sigma_Y's that catch!
Beware the Homogeneous Broadening, and shun
The Far-Off-Resonance Bandersnatch!

p.s. If you're on a job committee for one of my collaborators, I should point out that our sequences are just optimized for one kind of error, and of course by ignoring others the practical performance will be worse. Our paper should be thought of more as introducing new techniques/frameworks for producing composite pulses than as providing ready-made sequences that can be put into experiments. But please read it yourself if you're not sure.

01 December 2005

reading is fun! (damental)

Proof that if you make your slogan jarringly annoying enough, people will remember it decades later.

But seriously, reading is a good idea, and I should be doing more of it. Soon I'll try to blogs review of the (embarassingly short list of) books I've read this fall. But here's the quick summary.

  • Death of a Salesman by Arthur Miller. It's ridiculous how I haven't read or seen this before! This play is so moving that I wanted to clap after finishing Act One, even though I was just reading a script. And it's still incredibly relevant (if a little patriarchal): for example, you could argue that Requiem for a Dream just updated it for the 90's. And it's so much better than On the Road, which seems to be a lot more popular.

    The rest are in order of increasing specificity and convincingness.

  • Small is Beautiful: Economics as if people mattered by E.F. Schumacher. Classic 1970's critique of capitalism with a very Limits to Growth flavor. Argues that instead of prioritizing production through large-scale industry, we should focus on human fulfillment with smaller organizations. Nice ideas, but thoroughly missing is a look at the power relations that sustain the systems he criticizes. Without this, the whole thing starts to sound like those NYT editorials calling on the Bush administration to start respecting human rights. Plus the writing style often reminds me of management seminars and/or theology.
  • Economics and the Public Purpose by John Kenneth Galbraith. Classic 1970's defense of big-government liberalism. I would recommend it highly for the writing alone: crisp, witty, compelling and easy to read. I know the title makes it hard to argue this convincingly, but you really will enjoy reading this book. Anyway, I'll blog later about its contents in more detail, but for now will mention that its answers to the "why are things shitty?" question are much more specific and convincing than those in Small is Beautiful. It also proposes bold solutions that, while less convincing than the rest of the book (sometimes too radical, sometimes not radical enough, often dated), make a nice contrast with the usual alternative-less leftist whining.
  • The Wages of Whiteness by David Roediger. Explains how white working class racism emerged in 19th century America. Despite being much more academic and focussed than the above books, it's still very readable and engaging. I'll write later about how its specificity crushes Galbraith's arguments on the points where they clash. Also, almost all of its points are sadly still relevant today: how whiteness is constructed, why struggling separately against racial and economic oppression isn't likely to be successful, and how racism is so rarely about hatred. Again, the writing is inspiring; just check out the first few pages of the semi-autobiographical introduction (which you can do through amazon, or google book search) and you should be convinced to read the whole thing.
  • In the Shadow of "Just Wars": Violence, Politics, and Humanitarian Action. A collection of essays from Médecins sans Frontières staff, editted by Fabrice Weissman. Western political discourse usually misses the point so badly that it's offensive, especially when talking about poorer parts of the world. This book is an incredibly welcome counterpoint to that; but as it's late I'll keep this short. It discusses dozen humanitarian crises around the world, and critically reviews the responses to them. The results often aren't pretty, but I think we have a moral obligation to not to ignore them.
Wierdly, they all have a lot to do with money, though that wasn't my plan at the time.

Why have I read so little this year? I blame partly the thesis, but more the fact that I read too many blogs; for example, see this ironic blog post about how reading books is better.