Assorted links


Hey! I have a box of Peanut Butter Puffin Cereal on my shelf. It's really good.

Tyler, fyi a very important auction

Partial to Cinnamon Puffins myself, but this makes a lot of sense to try. There are many people who eat the same breakfast cereal meal after meal after meal -- as regularly as the Cubs don't win the World Series. And Barbara's is a brand in somewhat spotty distribution.

Breaking news. Your Willie-Nilly Barry, the fraudulent clown on duty, has done it again:

On # 1. My goodness. You are now searching for evidence in Science's backyards!

Can someone explain in simple terms how the stimulus could have cost jobs?

I understand how it could cost jobs in the long term, since it it financed by debt that presumably must be paid by taxes and the allocation of that money to pay for debt that paid for stimulus might cost more jobs than the stimulus money created.

But there have been no tax increase; we haven't begun to pay off the debt. And there's high unemployment. I simply can't see how jobs can be cost in the immediate short term by spending money financed by debt.

Think in terms of resource allocation. If we spent $499 billion on the stimulus, that transfers $499 billion in resources from somewhere to the stuff the stimulus was meant to prop up. Assuming it's all deficit spending, that's $499 billion from the bond market, that I'm guessing would otherwise have largely gone to the US private sector. The article suggests the private sector would have spent those resources in a way that employed far more people.

Also, they apparently spent $499,000,000,000 to save about 450,000 government jobs. That's comfortably over $1 million per job. Way to go, government. If you give me a million+ dollars a year, I'll create a lot more than one job with that, without even trying. Also, that's ignoring the million or so private sector jobs the stimulus destroyed through crowding out. Even the most optimistic projection they could put together said the government saved a net 659,000 jobs - we're still spending over $500k per job here. They could have given 6,590,000 people salaries over $50,000 with that money.

What I find more interesting, too, is that we have some theoretical reason to think those resources would have been invested productively in the private sector, in such a way as to reproduce at least as much was invested and probably significantly more. To break even, an entrepreneur has to get back at least what he invested, plus the cost of borrowing, plus taxes; a business that fails at this too long goes bankrupt. We have no theoretical reason to suspect that of government spending - in fact, the government might be spending resources on destroying resources, like, for example, paying prison guards to lock otherwise productive people up for nonviolent offenses, or paying regulators to help enforce a de facto cartel in the regulated industry, or spending billions on bailing out GM and having most of that money disappear. Sure, a lot of that money supposedly went to infrastructure, which might be a bargain considering the supposed positive externalities, but we have no way of knowing what the social return on such investments are. The stimulus's negative effects could be bigger than the GDPs of even some medium-sized countries.

I'd be curious to see how those who wrote about the positive effects of the stimulus, as well as T.A.R.P., like Zandi and Blinder, would react to this paper. Perhaps there's something that I, as a lay man that doesn't have the same sort of mathematical chops as an academic economist, just can't figure out and/or am easily missing, but it's hard for me to figure out how one side can claim large positive effects and the other side can claim negative effects.

Your second comment is of exactly the sort that makes me skeptical. The govt doesn't burn money. It doesn't pay its employees seven figures. So if your analysis makes it look like a govt job costs more than $1M, your analysis is wrong.

Your first comment sounds more plausible. Except when I just rapidly glanced at the paper, it looked like the analysis was performed by comparing state by state (or region by region) economic performance, noting different amounts of stimulus paid / received or something like that. And if so such an analysis should be completely insensitive to effects coming from straining the bond market, because that's nonlocal.

The reference to equilibrium theory strikes me as silly. It's quite obvious that at this moment in time, the private sector is not working optimally. It's of course not obvious what the fix is, and perhaps the fix is to do nothing at the govt level. But to use equilibrium theory as evidence in favor of doing nothing at this moment in time, this does not seem to conform to any logical implication of the word "evidence."

Let me see if I understand Conley & Dubor, the stimulus cost jobs because if the state & local governments had fired its employees they would have found jobs in the HELP private sector so because these government employees did not move to the private sector the stimulus cost private sector jobs.

Of course this ignores the possibility that if the private sector had actually been creating these jobs they just might have been filled by the unemployed private sector employees. Thus, they seem to assume the reason these imaginary private sector jobs were not filled during the great recession was because of the great shortage of labor.

Yeah, right!

Economics isn't like physics. You draw your curves first, and then you plot your data.

You forgot the operative word: DESTROYED!!

Only in an econ paper can a country DESTROY jobs with a stimulus used to enter into billions of dollars of contracts with private firms and save state and local government jobs.

#2. I'm generally extremely skeptical of purported essays on Christian music as they generally suffer from incomplete understating but ChristianityToday generally has quality articles and while someone passionate about music will always have minor quibbles with any expressed opinions about music, I thought that article was very good. It helped that it was not written by an outsider and that it was aware of its own limited perspective. I'm glad that it did not write about post-2000 rather than write about it poorly, yet at the same time I feel the last decade has been extremely interesting. But then, I've been a part of it, and this article felt like a good summary of what I've already gleaned from the decades I missed. Good choice on the informal writing style too.

To summarize #3:

(1) Over the past few years, states in a better budget position have seen more job losses than states in a worse budget position.
(2) The stimulus improved states' budget positions.
(3) Therefore, the stimulus caused more job losses.

I'm not sure the authors gave me enough to call (1) causality.

"You people are not making Christianity any better, you're just making rock 'n' roll worse."

Really though, I never thought I'd read an interesting article on Christian rock before.

Excellent post:
An environmentalist perspective on The Great Stagnation.

A must read.
The environmental lens is totally missing from the conventional conversation about economic growth etc etc

#1 You both missed the biggest "low hanging fruit" -- fresh water.

#5 - this is the kind of random link that makes MR so enjoyable... Keep it up!

Re #3:

Puffins are undervalued relative to penguins - marginalized, even. We can only hope a peanut butter grandma invoking cereal is the marketing coup they need to achieve a measure of social justice.


This has interesting implications for other languages whose economic vocabulary is translated more or less directly from English. The Mandarin term for inflation, "通货膨胀" or "swelling/inflating of the circulating currency" means something very different from "an increase in the overall level of prices in the economy." Because of this phrasing, the idea of inflation is strongly associated with increase in the money supply. The Chinese government would probably be unable to change the meaning to "an increase in the overall level of prices in the economy" even if it wanted to.

You beat me to it, Wimivo. The great Texan philosopher Henry Hill.

If Christian rock and praise music, with all of their shallowness, self obsession, and musical laziness, are truly our generation's contribution to a age-old tradition of music praising the Creator, then it speaks quite poorly of us.


IMO, both definitions should appear in a good dictionary. "Increase in price level" should be retained because it is now common usage. But, contra Reis and Sokolowski, "increase in money supply" is not just a "cause" of price inflation, because during a time of rapid technological advance, prices can decline even as money supply is inflated.

The two definitions have useful but distinct meanings. To throw one overboard is to impoverish language and degrade understanding.

On #3 (Conley and Dubor)

Has anyone actually read this paper?

On page 20 of their paper (Table 4), Conley and Dupor have a table that shows their main result: the number of jobs that they estimate to have been created or destroyed by the stimulus. In all private sectors, the point estimates are negative. But check out the confidence intervals in Table 4. With one exception, the upper limits of all the confidence intervals are positive. This despite the fact that they use 90% confidence intervals (instead of the usual 95%).

This means that Conley and Dupor's results are statistically insignificant. Formally, if we use their model to test the hypothesis that the stimulus caused a net increase in private-sector jobs, we will not be able to reject the hypothesis.
Or, to put it in English, what they have found is absolutely NOTHING.

There are numerous other problems with the paper such as their assertion that sales taxes are highly cyclical (WTF?)

But more fundamentally, state level studies tell us nothing about the macro effects of fiscal stimulus. It’s a near perfect example of the fallacy of composition. If the federal government builds a billion dollar military base in Fargo, North Dakota, I think most economists would agree (Conley and Dupor excepted) that the number of jobs increases in Fargo, North Dakota. Does the number of jobs increase at the national level? Possibly yes, but nothing in these state level studies actually addresses that question. It's the key reason why fiscal stimulus skeptics dismissed the findings of positive effects of fiscal stimulus on job creation in Wilson's paper, and Feyre and Sacerdote's paper. (And yet suddenly it doesn't matter in the case of Conley and Dupor's. Well as Emerson said "a foolish consistency is the hobgoblin of little minds.")

State level studies can’t tell us whether fiscal stimulus works just as state level studies can’t tell us whether monetary stimulus works. The Conley and Dupor paper is, simply the worst of a rotten lot. It only gets attention because of its eye popping abstract.

And that's the only reason why Mankiw posted it.

" This despite the fact that they use 90% confidence intervals (instead of the usual 95%)"

It's like a tell in poker.

The re-definition of inflation in the early 20th century did immeasurable damage to our understanding of political economy, even among experts. It is the single biggest step backward economics ever took.

Regarding #3: this unpublished paper appears to employ several logical non sequitors. First of all, their 90% confidence intervals for increases in government employment and decreases in private sector employment through September 2010 don't exclude the null hypothesis, yet they only mention this critical point in passing. Also, their core hypothesis, that continuing to employ government workers induces private sector losses even greater than the number of public sector job losses that would otherwise be accessible, makes very little sense in even a crowding out scenario. They also never explain the channel for crowding out to occur ( the standard channel for crowding out, treasury yields, approached record lows during the last few years), and never explain why there shouldn't be any time lag for crowding out to occur (do businesses check state government hires every morning so they know how many people to fire that afternoon?). They do admit "a crowding out story cannot be a full explanation," but then "leave completing an explanation for our estimates... as a question for future research." Obviously, this is a major cop-out. They never consider the alternative hypothesis that larger amounts of ARRA stimulus were supplied to the most hard hit regions, **because that's where they're most needed**, which could explain public and private sector jobs moving in somewhat negatively correlated fashion in every state. Finally, they claim their results are even stronger in the case where one only looks at employment through May 2010 instead of September 2010. However, one would expect under their crowding out hypotheses that once more government money is spent, more crowding out should have occurred. To find the opposite implies essentially that businesses anticipate the amount and rate of government spending, and pre-emptively cut back their spending in a disproportionate response even greater than the change in government spending. Of course, then it should work the other way around, too. That implies to me that the Ryan and Obama plans to cut the deficit by $4 trillion or so should result in an $8 trillion surge in private-sector spending any day now (sarcasm alert here). In short, their hypotheses are highly convoluted, poorly supported by their own data, and lead to fairly absurd predictions. Much of the recent and projected employment data can be explained by much simpler alternative hypotheses in some of the more mainstream papers they cite at the end.

To quote a couple of the co-author's former students: "Find another teacher!"

# 1 answer:,1518,757713,00.html

Mark A. Sadowski - Thanks for the quote. Next time how about you put it in quotation marks. ;-)

Peanut Butter Puffins are a tasty and not entirely unknown cereal product (see some of the other comments). What are you implying, sir?

That one struck me as weird as well. Peanut butter cereal is not exactly a novelty, and Barbara's is a long-established and fairly widely-distributed brand. I see some of their cereals in local Gristede's market, which is a pretty small and ill-stocked mainstream supermarket.

I personally prefer the regular Puffins, but the peanut butter ones are tasty. If I were going to go with a peanut butter cereal, I would favor Mother's Peanut Butter Bumpers.( You'd better really love them before buying a case of 14 boxes, though!

This shows which they last very much lengthier and thus saving you income which could otherwise are actually utilized to purchase new onesvbnvb.

Comments for this post are closed