Tyler Cowen’s The Great Stagnation: Government Spending Section

I’m quite taken with the central notion of Tyler’s new mini-book — that America has been picking the economic low-hanging fruit for decades or centuries, and that there’s a lot less of it around over the last three to six decades.

Before I get to the parts I like, I have to instantly respond to his section on government, because it has some flaws and falsehoods that I’m really stunned to hear from him — because I know for a fact that he knows they’re not true.

He uses the marginal utility argument (you only need so many apples, so additional — marginal — apples yield progressively less benefit per apple) to argue that as government grows, the marginal additions yield progressively less value:

Even if you think everything our government does is awesome, successive units of government are on still on average less valuable than the core functions.

He’s basically applying his low-hanging fruit contention (which is different from the single-good marginal-value argument) to government: all the high-value services (i.e. the low-hanging fruit) have already been delivered. That contention may be true; more below. But first we need to look at the basic facts he presents.

Full credit: though he waits a bit to do it, he acknowledges that government expansions of medicare, social security, etc. are not necessarily subject to the marginal-value rule, or at least that they shouldn’t be considered in evaluating it.

The relevant number for government here is not “government as a percentage of the economy,” because that includes a lot of transfer and welfare and social security payments, which simply shuffle money from one person to another. A better measure is “government consumption” – what government itself is doing

In addition to the money-shuffle argument (which is correct): medicare etc. are not so much additional services as existing services for which the money is funneled through government instead of through private enterprises. (Far more efficiently than the private enterprises, it so happens, so the value returned from those additional government revenues/expenditures is higher, not lower, than if the money flowed through private enterprises.*)

Here’s where it gets really dicey.

that figure [government consumption] commonly falls in the range of 15 to 20 percent of U.S. GDP. As long as the absolute size of government consumption is rising — as it generally does — we are getting less value than our measurements indicate.

What in the heck???!!! He cites consumption relative to GDP, states that it’s been pretty steady, then talks about the “absolute size” of government consumption, and says that it generally rises. He really seems to be talking out of both sides of his mouth.

1. Yes, the absolute size of government consumption does tend to rise — no duh. The population is getting larger and the economy is getting larger even faster.

2. The relative size, on the other hand — which really is the most useful measure, which is presumably why he cites itdoes not generally rise (which he acknowledges) — making his “it generally does” a patently false statement.

Update! See corrected graphs here. The Federal and State/Local labels should be reversed.


(Expenditures: NIPA table 3.9.5, lines 12, 17, and 22; GDP: NIPA GDPA.)

I don’t even have to comment on this graph; the import is clear on its face. Tyler’s doing the typical righty thing here: making assertions based on facts from the sixties and seventies — facts that have not been facts since that time.

And I know Tyler knows this.

But to return to his basic assertion, which is not and really can’t be based on quality data: has government already picked all the low-hanging fruit? We can’t really answer that for the reason that he states (accurately): in measuring GDP, government services are measured by what they cost, not what people payed for them. They could be worth more than they cost, and they could be worth less. Since we don’t have the pricing information by which to judge their value, we have to use other methods that are probably far less accurate, and are certainly more subject to disagreement.

Are additional government services inevitably less productive than existing ones? Let’s think about the government consumption expenditures that were thrown into enforcing civil rights and fair employment starting in the sixties. Were those less productive than consumption expenditures that preceded them? I would argue “no,” and suggest that just because some government public service has not been offered in the past, it is not necessarily less productive than existing services (though it may well be). It’s an assumption, not anything like a fact or a given.

And in fact, it’s not even asking or answering the right question, which is: is money taken from circulation in the private economy and spent by government on service X more or less productive than if the money was left circulating in the private economy? With $55 trillion of purely financial assets in this country — most of which is chasing its own tail and producing nothing — and desperate, gaping shortages of clearly productive government spending, I would say “yes.”

The basic truth, though — by Tyler’s own assertion — is that we just don’t know. Which means that Tyler could have deleted this whole government section from the book without removing any of the book’s value. He just interjects some ideological axe-grinding here — promulgating a belief that is at best faith-based, and at worst completely false — even though it has little or nothing to do with his central thesis. This government section is a digression — an unhealthy pseudopod that could be lopped off with great benefit to the creature as a whole.

* 25% of health-care dollars disappear on the way through the private-insurance funnel. Only 3% of the dollars disappear on the way through Medicare. Health care: 17% of GDP. 22% of that is 3.8% of GDP that we could use for other things if everyone was on Medicare instead of private insurance. That’s half a trillion dollars a year, which would pay for almost all our governments’ nondefense consumption expenditures combined.


Posted

in

, , ,

by

Tags:

Comments

12 responses to “Tyler Cowen’s The Great Stagnation: Government Spending Section”

  1. jazzbumpa Avatar

    So the guy’s a bare-faced liar.

    Tell me again why I should read him.

    Cheers!
    JzB

  2. Tyler Cowen Avatar
    Tyler Cowen

    You are misattributing a position to me. Pre- and post-1973 — the fundamental comparison of the book — there is a real difference.

    1. Asymptosis Avatar

      Tyler: “You are misattributing a position to me. Pre- and post-1973 — the fundamental comparison of the book — there is a real difference.”

      I don’t understand. There’s a difference between what? Rates of growth in govt consumption spending as a % of GDP? That’s the point I was making — it hasn’t gone up since then — and objecting about — you suggest that it has/does.

      Or something else?

      And what position am I misattributing? Truly, don’t understand.

  3. jazzbumpa Avatar

    Back to what changed: In the Fed non-def category, look at that inflection point at 1975/6. Also note at that time the tricolor total pretty much levels off. Before then, from WWII on the defense component was shrinking – a trend disguised by the blip for Korea in the early 50’s.

    Stacked area graphs are illustrative, but for categories not at the bottom, it’s hard to discern changes over time. Has state and local remained constant? Looks like it, but with that wavy snake, it’s hard to be sure.

    Cheers!
    JzB

  4. Asymptosis Avatar

    One problem I think I see here: Tyler conflates two issues:

    1. The marginal utility of government services as a single “good” (as you get more, the marginal utility decreases, as with apples), and

    2. The low-hanging-fruit effect, which relates to multiple, different, additional government services or “goods.”

    They’re similar and sort of related, but quite different.

  5. […] in an email pointed out a whopper of an error in a previous post, in this […]

  6. steve broback Avatar
    steve broback

    Inflation adjusted government consumption per capita is also representative — and certainly not flat. “And I know Steve knows this.”

    “Are additional government services inevitably less productive than existing ones?” Yeah, pretty much. Have you ever responded to the Gruber quote here?

    66 out of 71 studies enough for you?

    http://stevebroback.com/2011/the-streams-cross-obamas-numbers-slip-below-reagans-for-good/

  7. […] This would appear to handily refute the non-intuitive argument Krugman made (and bloggers parrot) about the imagined efficiency of Medicare and […]

  8. Asymptosis Avatar

    — Inflation adjusted government consumption per capita is also representative — and certainly not flat. “And I know Steve knows this.”

    Who are you quoting? I actually have no idea what that graph would look like, would love to see it. Certainly important. I’m thinking, since per-GDP is flat, it would go up with productivity. ??

    — “Are additional government services inevitably less productive than existing ones?” Yeah, pretty much. Have you ever responded to the Gruber quote here?
    66 out of 71 studies enough for you?

    Have you read the Mueller paper? Is it any good? In any case according to Gruber it doesn’t even address the question asked here.

    I’m constantly at a loss, though: if you feel this way, why do you keep saying that china’s so much better than us, when their businesses, by capitalization, are 90% government controlled?

  9. Asymptosis Avatar

    @steve broback

    I also wonder: do those studies look specifically at financial services, or split them out? Those are the big services government provides in the U.S. (pensions, health insurance), and it seems to do that far more efficiently than the private sector. Seems like there might be a key distinction there.

  10. tin Avatar

    Hey would you mind stating which blog platform you’re using? I’m going to start my own blog in the near future but I’m having a hard time deciding between BlogEngine/Wordpress/B2evolution and Drupal. The reason I ask is because your design seems different then most blogs and I’m looking for something completely unique. P.S My apologies for getting off-topic but I had to ask!