Paul R. Krugman, formerly of Princeton and now of the New York Times, has been laying into the Sanders campaign for junk economics, because of a paper by one Dr. Gerald Friedman, which projects unusually high economic growth if Sanders’s entire program were passed.
But now we get input from another respected economist. James K. Galbraith called Drs. Krueger, Goolsbee, Romer, & Tyson to task for not actually running the numbers on Friedman’s analysis. Apparently Gerald Friedman’s much-derided growth projections for the whole Sanders framework are, well, entirely consistent with current economic science.
So we have a handful of high-profile wonks who dismissed the work out of hand because it sounded implausible.
Oh, it’s a short PDF, & I don’t want to quote too much of it, but this is a good line:
Thank you very much Mr. Guinea. It’s a shame how the right-wing propaganda train rants its ignorance and dogma; and especially a shame how many “moderate progressives” fall for the same dogmas, unwilling to dream big.
Whether he makes it to the Oval Office or not, I hope Bernie Sanders leads a revolution in American discourse. (Maybe we’ll see a Sanders disciple get the Oval in 2024.)
The Friedman model was cooked. Galbraith misleads his readers. This isn’t a case of dreaming big. Passing single payer is dreaming big. Claiming that passage will have one consequence or another is something else.
In a blog post Kevin Drum dug a little deeper. It’s not just one number. [INDENT]Productivity growth will double compared with Congressional Budget Office projections [according to Friedman’s analysis]—and in case you’re curious, there has never been a 10-year period since World War II in which productivity grew 3.18 percent. Not one. And miraculously, the employment-population ratio, which has been declining since 2000 and has never reached 65 percent ever in history, will rise to 65 percent in a mere 10 years.[/INDENT] And it will be much harder to drive the emp/pop ratio ever higher given the aging of the population.
I was going to create a chart like this, but Krugman did it for me. It shows historical growth rates relative to those projected by Friedman over 10 consecutive years. In this case the figure is GDP per person.
This isn’t dreaming big. But I don’t think it’s imitating the conservatives either. It’s a tacked on analysis of an aspirational plan, endorsed by a policy director in the midst of a political campaign. Note that the economics of the plan itself hasn’t been challenged by Krugman et al. Just one analysis of it.
At any rate, you can see why the members of the Council of Economic Advisers didn’t offer an analysis of their own. That takes time. And really it’s unnecessary: the burden is on Friedman to show the plausibility of his analysis. I seriously doubt he can.
What I’m reading is that the 5.3% growth is not meant to be a ten-year mean, but a high in the first two years. So Krugman and the CEA four are arguing against a distortion of the paper’s conclusions, not the paper itself.
ETA: Wait, no, apparently that’s spin. I’ll have to track down the actual paper.
ETA2: Wow, that PRK graph is…definitely spin. Of course we’re not going to have 4.5% growth over 57 years.
I’m still looking! Found a whole mess of links to the ‘kerfuffle’ (that is the technical term) here:
It sounds like Gerald Friedman ran Sanders’s plan through some conventional models & got some numbers that do fit the pre-2008 growth trend but don’t fit certain other economists’ assumptions about, um, structural change in what full employment looks like. So the actual argument among economists is about what “full employment” can mean with an aging society.
This is a tempest in an economic teapot, arguing about structurally low labor participation, and really not that important to whether Sanders’s plan is better or worse than anyone else’s.
When Sanders’s policy director (whose background is poli sci, not econ) heard about the study, he said, “Wow! Thanks for your endorsement!” (paraphrase) and this turned into a…kerfuffle.
Then Gerald Friedman said, “Actually I’m voting for Hillary, for other reasons. Growth is just one issue; there are others.” (paraphrase)
People don’t seem to understand how gigantic a number 5.3% growth is for the American economy. Just hearing the number should make anyone scoff. It should have been an immediate trigger. Wait. I got 5.3% for my result? I must have done something wrong. Well, my calculations aren’t wrong so the numbers I fed into them must be.
That was my reasoning, and the pdf from Jamie Galbraith says it presumably was the reasoning of others. (I can’t reconcile the numbers in the chart in Measure for Measure’s link with the numbers Galbraith quotes, so they must be different measures of different things.)
Possibly Sanders’ “plan” is so gigantic that it would transform the country if enacted by a snap of the fingers. It won’t be. No gigantic plan is or ever has been. (The New Deal never came close to what many reformers wanted, even before so many pieces of it were rescinded or scaled back.) What is the analysis if only bits and pieces get enacted? What is the analysis if everything is enacted at a smaller scale? What is the analysis if a multi-year fight is necessary to get it passed?
Sanders’ plan is fantasy in any real-world setting. Any. Even one with a fully Democratic Congress. With a hostile Congress it is lying with statistics. Dangling a 5.3% growth rate (and it can’t be a brief spike, nor does Friedman say so anywhere in his article) before people is pandering at the same level as building a wall that the Mexicans will pay for.
The technicalities of the analysis is not the issue. One can always enter numbers to make any desired outcome occur. The issue is the validity of the numbers entered. These numbers don’t have any.
OK, there’s thispair of articles for Dollars & Sense. I’ve only skimmed it. I can understand the argument for temporary accelerated growth, but the shocking drop in the poverty rate looks counter-intuitive to me.
This may have been a half-assed analysis, if you’ll excuse the expression.
But if the (apparently pretty normal) models and assumptions Friedman used overstate the effects of the Sanders tax & redistribution plan, then presumably they would also overstate the effects of whatever plan a more mainstream Democrat puts out. That implies to me that if Sanders’s plan is not going to get this big an effect, any more mainstream plans will have effects on growth and poverty nearer zero than advertised.
That thought is not comforting, if you expect to elect a mainstream Democrat.
I see what happened. It’s linked to in the HuffPostarticle you linked to. Click on the link under “an analysis”. It’s a pdf so I can’t link directly to it.
If economists’ standard models work anything like physicists’ do, then you can’t feed unprecedented huge inputs into them. In order to make a model usable, you need to assume that some terms are small, and that higher orders of those terms are smaller, and that at some point they’re negligible. If they’re large enough that the things that you’re neglecting aren’t negligible, then the model simply doesn’t work.
Well sufficiently fast growth will boost tax revenues and reduce welfare payments, therefore lowering the budget deficit. So it’s a convenient feature of the model.
Growth rate of per-capita GDP, 2016-26. That’s where the 10 year figure comes from. And here’s a quote from page 12 of the paper, “The Sanders program… will raise the gross domestic product by 37% and per capital income by 33% in 2026; the growth rate of per capita GDP will increase from 1.7% a year to 4.5% a year.”
It’s a 10 year figure. Galbraith appears to have misrepresented the analysis. And Exapno has it correct: when your model spits out something ridiculous you re-calibrate.
That’s how I would spin it. I think the study was a hack-job and Galbraith’s characterization was worse. But I’m not sure I’d even call it a kerfuffle given that the candidate himself hasn’t mentioned the study to my knowledge.
Note that Dean Baker of the liberal think tank EPI thinks the Friedman study was out of line as well. If that helps.
