I got 99 problems, and they all that I spilled coke on my laptop

So, you know, no more blogging until I do one of:

1) figure out if my magical Costco warranty covers spill damage and will replace my computer


2) trick my computer into booting Linux from the CD drive before it reaches the part where it realizes it no longer has any idea how to boot Windows.

PS –

Maybe I’m just a baby.


Simulating Radicalism: A Lesson from Paul Krugman

Paul Krugman wrote today about “the political economy of redistribution.” The title is important. He isn’t just talking about economics, you guys, he’s talking about political economy. In fact, for Krugman,

The real question – arguably the central question of political economy – is how much to redistribute. And it’s both interesting and important to try to understand how that decision gets made.

This reasoning suggests an incredible misunderstanding of what political economy actually is. It isn’t, as Krugman suggests, a constrained optimisation problem in which an economy realizes decreasing returns to redistribution scale and increasing marginal costs, whatever those are, of a more equal distribution of income and opportunity. Were that the case, “political economy” would be a particular math problem that senior high schoolers do in their first semester of calculus, and how much to redistribute would be a math problem. This consequence is unacceptable, and there are more reasons to reject Krugman’s definition.

To begin with, political economy has historically meant the particular social arrangements that constitute the mode of production. Questions asked include “who owns the means of production?,” “how are decisions in firms made?,” and “what institutions characterize the social structure of accumulation?” Redistribution is present in the last question. Inasmuch as the tax and transfer system claims wealth that would flow to the capital class and redirects it to the working class, it is one determinant of the social structure of accumulation, but note that it “one determinant of the social structure of accumulation” is a long way from “arguably the central question of political economy.” The claim doesn’t even work with the weasel word “arguably” included.

Krugman’s focus is the use of a particular institution — the tax and transfer system. His question is efficiency, and the answer lies in seeking out the most effective means for Democrats and Republicans to make their narrative of redistribution work. He writes

operatives on the right are clearly worried that their three-decade run of success [convincing the median voter to vote against his/her interests] may be coming to an end. Indeed, the whole panic about the lucky duckies and all that can be seen as reflecting a great fear on the part of the right that any day now the median voter will realize where his true interests lie, and start supporting much more redistributionist policies.

But this is distinctly not a political economy question.

Were I in Krugman’s place, I would be tempted to locate my work in the “political economy” field rather than the “economics” field. Economics, as a social science, has a tendency to ask some boring questions, while political economy allows questions about the drama and grandeur of modern politics. Besides, “political economy” has a nice historical mystery about it, and even mentioning it pisses off at least one editor at the WSJ editorial board. However, “how will politicians on both sides convince voters that their narrative better serves voters’ economic interests than the other side’s?” is not a political economy question despite its being a question about both politics and economics. Changes in the tax and transfer system, even radical changes in the tax and transfer wouldn’t change the institutions that characterize the social structure of accumulation inasmuch as the institutions themselves wouldn’t be any different. The only change would be in the degree to which the capitalist class is able to redistribute income upward in brief spans of time.

So, you know, Paul Krugman: still a pretty un-radical dude.

Noah Smith characterizes being on the liberal blogger team this way:

I’m a “liberal economics blogger”. I guess. Meaning A) I’m a liberal, and B) I’m an economics blogger. But also meaning that I stick up for Paul Krugman, and that I spend a lot of my time on this blog criticizing the dominant paradigm in economics, which currently is a conservative paradigm.

But what real criticism has either of them done of the dominant paradigm in economics? Were we to borrow from Lakatos and characterize the hard core, positive heuristics, and negative heuristics of each side, what differences would we see? Sheila Dow (with the help of Weintraub 1985a and Backhouse 1991) characterizes the hard core of neo-Walrasian general equilibrium — the mainstream — theory this way1:

HC1 There exist economic agents

HC2 Agents have preferences over outcomes

HC3 Agents independently optimise, subject to constraints

HC4 Choices are made in interrelated markets [6]

What part of this hard core has Krugman or Smith critiqued? The criticisms of really existing capitalism from Smith, Krugman, and the liberal blogosphere generally have to do with the failure of the real market to live up to the theory, but, putting any lags, leakages, and information asymmetries in the “subject to constraints” category, these criticisms do nothing to attack “the dominant paradigm in economics.”

Krugman’s cheap grab at outsider status from his “political economy” work should be taken with a grain of salt. There is no reason to believe that Krugman or any other large readership liberal blogger else takes seriously the idea of a non-capitalist economy, and this means that there is no reason to believe that Krugman or any other large readership liberal blogger deserves any credit for attacking the dominant paradigm in economics.

1 – Dow, Sheila. Foundations for New Economic Thinking. Palgrave McMillan, New York: 2012.

Friedrich Engels at Naked Capitalism

Dan Kervick has, I think, done the best job categorizing QE3, calling it “shamanistic economics.” He wrote:

Here’s how it works in theory: Suppose there is something X the Fed would like to see happen. The Fed is supposed to make X happen by announcing that it intends to make X happen and that it is doing some other thing Y that is aimed at making X happen. The hope is then that a significant number of people thinkthat there is a causal connection between Y and X, and that Y causes X… What is important, the defenders of this approach say, is that people believe quantitative easing will cause a stronger recovery. They will then start to invest in production, hiring and consumption more readily because they are expecting improved conditions… It doesn’t really matter what quantitative easing would have accomplished on its own if people didn’t have these beliefs.

Especially in light of Arin Dube’s skepticism that Bernanke and the Fed actually has any ammo left to push up inflation at the zero lower bound and the effects this skepticism should have on inflation expectations, “shamanistic economics” nails what the expectations fairy is supposed to facilitate. So, as a term for thinking about what exactly QE3 is, shamanistic economics  does a fine job.

I have a totally separate problem with this article: the idea that shamanistic economics is for some reason illegitimate. Kervick seems to believe that it’s possible to consider something like quantitative easing in a vacuum. Above, he mentioned “what quantitative easing would have accomplished on its own” (emphasis added). He later writes

All that matters is that apart from whatever actual causal connections exist between X and Y that operate independently of expectations, there are also a lot of popular beliefs about the connection between X and Y that cause people to act with the expectation that Y will cause X. (emphasis added)

What he’s after here is what the collection of actions called quantitative easing would do to an economy in which people’s expectations didn’t change, or in which people’s expectations didn’t affect economic outcomes. That goal is absurd and aligns almost identically with what Frederick Engels meant when he spoke about the scientist’s seizing “nature without any foreign addition,” and the same criticism should be made. As Francois George writes, “Far from preceding the scientific attitude, reality and the object were constituted by it,” which is to say that without a ready-made body of scientific theories and community of scientists, reality and the object would both be unintelligible (80, paid, sorry).

The same objection applies to economic policy/experiments  — any economist claiming he has a good model of how the economy works outside of how people would react to it doesn’t have a good model of the economy works. This is, in a way, a post-modern extension of the Lucas critique. As Simon Wren-Lewis explains,

The classical example of the Lucas critique is inflation expectations. If monetary policy changes to become much harder on inflation, then rational agents will incorporate that into the way they form inflation expectations. A model that did not have that feedback would be ‘subject to the Lucas critique’.

In this description, the policy happens prior to the expectations adjustment, which is to say, unless we assume perfect and immediate adjustment everywhere, this is the chain of events: policy-makers make a real/objective change in economic conditions; real changes occur; expectations change; feedback from the expectations changes causes other real effects.

Now, though, the Fed can’t make real stimulative changes (though it could, presumably, ruin everything forever by making real contractionary changes). At the ZLB, its ability to cause inflation has been neutered, and it can no longer lower nominal interest rates. What the latest round of QE relies on is short-circuiting the Lucas critique. Rather than making real changes and arguing that the expectations feedback won’t undo them, the Fed is skipping the making real policy changes step. The Fed is skipping what’s supposed in a rational/objective universe to be a sequence of cause and effect, and is instead relying on subjective interpretations of future economic conditions to bring about those conditions.1

Kervick thinks this plan is stupid because it shouldn’t work in something like an objective economic reality. However, inasmuch as economic reality is comprised of people with expectations and that, without these people, “economics” as such doesn’t exist, the impossibility of proving the effectiveness of further easing in an objective economic reality is unconcerning.

This isn’t to say that the Fed policy should work. Even if unemployment turns around tomorrow, it will be unlikely that that change was in a strict sense “caused by” the Fed action.

The success of the Fed’s actions, Kervick writes, “depends on the perpetuation of false and superstitious beliefs among the public,” but, as roguishly as I could, I’d counter sure, but not “false” in the sense of “not true,” but of a different kind.

1 – For a literal, literary, and hilarious example of what this looks like in magic, see Jonathan Strange & Mr. Norrel, chapter 22

Thomas Friedman Believes in Thomas Friedman’s Previous Explanations of the World

Every time I read a Thomas Friedman article, I have to compare it to the Thomas Friedman op-ed generator at McSweeney’s. Today though, formulaic writing wasn’t Friedman’s problem. His problem instead was that his narrative doesn’t work.

Friedman’s basic argument is a flat world justification of pursuing higher education of some kind and focusing stimulus efforts on increasing educational access and attainment. He writes

Because when Clinton first employed his phrase [“work hard and play by the rules”] in 1992, the Internet was just emerging, virtually no one had e-mail and the cold war was just ending. In other words, we were still living in a closed system, a world of walls, which were just starting to come down. It was a world before Nafta and the full merger of globalization and the information technology revolution, a world in which unions and blue-collar manufacturing were still relatively strong, and where America could still write a lot of the rules that people played by.

That world is gone. It is now a more open system. Technology and globalization are wiping out lower-skilled jobs faster, while steadily raising the skill level required for new jobs. More than ever now, lifelong learning is the key to getting into, and staying in, the middle class.

So, as a result, like Estonia, we should train our kids better, which means educate educate EDUCATE. Instead of writing this article, he could have tweeted “You guys, skill-biased technical change is a thing,” but, you know, oh well.

The policy prescription is something I’m on board with, but Friedman’s rationale is suspect. First, the shift from “work hard and play by the rules” to “work harder, regularly reinvent yourself, obtain at least some form of postsecondary education, make sure that you’re engaged in lifelong learning and play by the rules” doesn’t indicate that “work hard and play by the rules” no longer applies. It does indicate that the rules have changed, if something called “the rules” ever actually existed, and that it’s still important to work hard and play by them. Working hard just happens to include regular self-reinvention and a more secure place in the education arms race.

What’s really suspect though is Friedman’s exegesis. The logic is something like this: foreigners are well-educated now, and careerism requires harder work than before, so it’s probably true that careerism requires harder work because foreigners are well-educated now. People need to know about this, and urgently! If you’d written a book called The World is Flat though, this narrative of international competition forcing Americans to work harder would be extremely appealing. If it were accurate and skill-biased technical change plus globalization were making low-skill workers extremely uncompetitive, you’d probably think you understood the world economy pretty well. Since 1992 is Friedman’s new world date, I looked at some BLS data from then to 20121.

First I looked at employment levels to figure out if the composition of working Americans in the workforce is changing.

It is. Indexing from the 1992 employment level, holders of a four-year bachelors have added about 75% of the jobs they had in 1992, while the other two groups have declined slightly over the period. Score one for Friedman, kind of. Endpoint comparison says that there are fewer jobs for high school grads and those without high school degrees in 2012 than there were in 1992, but it would be extremely difficult not to notice the timing of the biggest decline — the current recession — or that several times post-1992 the employment level returned to or was above its 1992 level for both low-skill groups. Regardless, the percentage of the total workforce with a college degree is obviously increasing.

Second, I looked at employment to population ratios…

…which didn’t reveal very much. It’s probably true that the number of people 25 and older with less than a high school degree is declining, which may explain the slight upward trend in that group’s employment to population ratio from ’92 to 2008, but I’m more confused about what’s going on with the college degree group. These are the people who, if you buy Friedman’s argument, should be more in demand and more insulated from the rest of the world’s development, but instead, job growth for this group hasn’t kept up with the population change, and not only in this recession. For some reason, while the employment/population ratio hadn’t fallen below 78 percent in the period from 1992 to 2000, the early 2000s saw its range shift from between 78 and 79% to hovering around 76%. What exactly is going on here I’m not sure, but Friedman definitely didn’t score a point. Finally, I looked at the unemployment rate for these three groups and compared them to the national average.

This is where I expected to be able to find something serious. Since the unemployment rate shows only those who can’t find work who are looking for it, Friedman’s flat world interpretation suggested something obvious: an open world should penalize low-skill groups and reward high-skill groups, or, since the open world means firms don’t have to hire low-skill American workers and can instead hire high-skill foreign workers, low-skill workers should have trouble finding employment at times high-skill workers don’t. As a result, the advantage of having a college degree should be widening. Instead, it’s varied smoothly between six and ten points in the last two decades, and is right back where it started in 1992. What’s driving unemployment for each group doesn’t seem to be educational attainment in the rest of the world, but rather the state of the national economy.

The point here isn’t that there aren’t advantages to education. Rather, the argument is that the employment penalty of not being educated is about the same now as it was in 1992 and hasn’t changed significantly in that period. Note that I’m ignoring wages partly because I have had about enough of navigating the BLS website for today.

I have a theory (it is not an original theory) for why Friedman’s predictions haven’t punished the low-skill group as he says it should — high-skill international workers aren’t competing with low-skill domestic workers. The world high-skill group’s population has grown significantly and often provides abstract services, which can easily be outsourced, while the domestic low-skill group’s population has shrunk and often provides concrete, immediately present services that can’t be outsourced. The low-skill group isn’t penalized because it isn’t competing with the high-skill group. I don’t know where to look for data on that though.

1 – All data are from the BLS Current Population Survey. If anyone knows how to trick the BLS website into graphing several series in the same space, please let me know.

Throwing in the Towel (actually)

Bruce Bartlett probably isn’t going to vote because he doesn’t like either candidate. That… kind of makes sense. Here’s why:

The usual response to this kind of childish demand for a good choice — “I want exactly what I want and not a cheap substitute!” — would be something like “Even if you don’t like either candidate, there’s probably one you dislike less.” That doesn’t work though. It may, in this election, keep the greater of two evils out of office, but election success is (read as) an endorsement of a party’s platform. That is to say, one option is supporting a GOP platform that believes inflation transfers money to Wall St. (then why do banks hate it?), that the dollar should be anchored to something, that we need more protectionist trade policy, that the Electoral College and voter ID laws are necessary measures to protect us from fraudulent elections forever, and privatizing internet regulation. But really that’s all bonus absurdity, since they’ve somehow convinced themselves that unborn fetuses are somehow covered by an amendment that begins “All persons born or naturalized in the United States” (emphasis added, of course). The other option is a Democratic Party platform that mistakes “mutual interest and mutual respect” for a description of future policy in Iraq, doesn’t mention drones (here’s info on Pakistan – note the shrinkage of unknowns and civilians after we started designating everyone who is near a terrorist a terrorist), brags about sanctions against Iran like they’re an accomplishment, and, in something that’s preemptive strikes away from the Bush Doctrine, focuses on “a prosperous and inclusive economy, our unsurpassed military strength, and an enduring commitment to advancing universal values.”

Voting for the lesser of two evils necessarily makes things worse. There’s no line item exemption for particular planks of a party’s platform, so what’s a conscientious voter to do?

Throwing in the Towel (kind of)

Paul Ryan lies a lot. That’s fine, really. I don’t know why we expected him not to, but maybe the specific things he lies about are what’s concerning. It’s frustrating when he lies about things that happened in the past, like the results of the Bowles-Simpson commission, or how much blame Pres. Obama deserves for a certain factory’s closure, or his marathon time (because that’s important). To say that things that have already occurred happened in a certain way when, in fact, they didn’t, is cheating. But his lies about the future? For instance, when he makes particularly claims about the effects from his budget and then admits that he and Romney haven’t run the numbers yet?

And Ryan’s not the only one paddling this canoe. When Romney released his budget, the Tax Policy Center, according to Ezra Klein, “bent over backwards” to make the numbers add up to what Romney claimed. It’s bad news, probably, for your budget when

Our major conclusion is that a revenue-neutral individual income tax change that incorporates the features Governor Romney has proposed – including reducing marginal tax rates substantially, eliminating the individual alternative minimum tax (AMT) and maintaining all tax breaks for saving and investment – would provide large tax cuts to high-income households, and increase the tax burdens on middle- and/or lower-income taxpayers. This is true even when we bias our assumptions about which and whose tax expenditures are reduced to make the resulting tax system as progressive as possible. (emphasis added)

ends up in the TPC’s final analysis (page 2).

The implicit argument here is that Romney and Ryan haven’t given us enough information about what they’re planning on doing to make any of their plans make sense. I question whether that matters.

Inasmuch as we’re still super-concerned about growth — we are still super-concerned about growth, aren’t we? No one’s happy that we’re back to the pre-recession trendline’s slope but at a level significantly below the former trendline? — we’re still talking about stimulus, just of the budget variety, rather than of the ~trillion dollar spending bill variety.

The TPC, in a blog post in early 2009, proposed a word of caution:

Here’s a dirty secret about economic stimulus: We’re making a lot of this stuff up. It’s based on a combination of often inconsistent theory and ambiguous empirical evidence.

This caution has been forgotten in the certainty that Delong and Krugman express from the left and that Mankiw, Hubbard, et al. have expressed from the right. Having failed to predict the crisis in which we’re now mired, everyone with a model claims accurately to be able to predict the results of his own policy prescriptions. This is obviously absurd.

The TPC post proposes one option: use the states as experiments. They proposed treating all states that start with letters A-F with one policy set, G-L another, M-P a third, and so on, to find out what exactly might be more effective. For this to work, the balanced state budget requirement would have to be put on hold, which doesn’t faze me, but might bother some people. Labor mobility, natural resource endowments, and pre-experiment state of development might cause problems, but statisticians can take care of those things. The goal would be to collect more data at a particular point in time, and thus get a better idea of what works and what doesn’t.

That’s reasonable, but maybe silly. At a certain point, the purpose of political power became making good economic decisions. It’s important to remember, though, that no one has day-to-day problems with the national debt, while people do have day-to-day problems with bills, food, etc., and that should redirect our focus, say, this way:


Political power can be used to make those at the bottom end of the distribution better off, which is a good thing regardless of whether it encourages economic growth, and it should be noted that having 1.5/10 people excluded from the potential consumer class probably doesn’t increase economic growth.

The question is about priorities. We don’t know with a high degree of certainty what will stimulate growth, broaden the tax base, decrease the national debt or at least be deficit-neutral, “get America back on track,” etc., but for some reason those claims are a part of every budget. Meanwhile, the Dutch model provides a set of programs that do a good job keeping the worst off afloat (for an overview of these programs, see here), but no one has the courage or, more likely, the political capital to say that policy doesn’t need to concern itself with people who are already fine. Political power, in this latter view, should first be used to remedy the situations of those who are worst off and should concern itself with growth and budget-balancing only secondarily.

Of course, reprioritizing thus would require Congressional term limits to create a term in which it makes sense for an individual legislator to spend whatever remaining political capital s/he has to do good things, so, you know, don’t hold your breath.