Nick on Chinese stimulus

He emails me:

Why isn’t China post-2008 considered a great real world experiment of Keynesian stimulus?
-It was large enough (roughly 3x US stimulus as % of GDP iirc)
-The govt was starting from relatively low levels of public debt (reported anyway) so there wasn’t a crisis of confidence issue as in Europe today
-It was invested heavily into infrastructure and projects (like solar) slated to improve long run efficiency in an economy that arguably had room to increase capital stock per capita
-The political leadership directing the stimulus spending is made up of engineers, scientists, and other highly educated politicians who did not need to run for public election and thus could focus on longer run trade-offs
-The goal was to bridge a period of weak external demand rather than fundamentally alter the economy
Is this not a Keynesian’s dream scenario? My guess is that now Keynesians would say the spending was misguided and wasteful. But that’s the point! Western elites constantly praise Chinese leadership for their acumen — even today there’s an incredible amount of faith inside and outside the country that they will thread the needle — but they too were wasteful when it came to rapidly expanding government spending.
I have greatly enjoyed the recent pieces on China as I currently live in REDACTED…Fwiw, my take is that yes, the 2009 Chinese stimulus was a great test of large-scale govt stimulus and that we are seeing the results of what this looks like *in practice* rather than *in theory*. In practice, large-scale government stimulus is an invitation for corruption and a diversion of resources that builds up knowledge and capital in unproductive areas. As Obama said after the US stimulus package: there’s “no such thing as shovel ready projects.” Even in China.


'In practice, large-scale government stimulus is an invitation for corruption and a diversion of resources that builds up knowledge and capital in unproductive areas.'
Strange - I thought that essentially all large Chinese projects, considered stimulus or not, were directed by the government. Which would lead to the not exactly uncontroversial position that large scale government spending is an invitation for corruption and a diversion of resources.

Witness the military-industrial complex - Eisenhower certainly seems to have correctly predicted that the American military build up of the last two generations would result in the build up knowledge and capital in unproductive areas - 'In the councils of government, we must guard against the acquisition of unwarranted influence, whether sought or unsought, by the military industrial complex. The potential for the disastrous rise of misplaced power exists and will persist. ... Partly because of the huge costs involved, a government contract becomes virtually a substitute for intellectual curiosity. For every old blackboard there are now hundreds of new electronic computers. The prospect of domination of the nation's scholars by Federal employment, project allocations, and the power of money is ever present and is gravely to be regarded.'

And, Ronnald Reagan, former Governor of California, new that the stimulus California needed was military aircraft, missles, and star wars to help along the California Miracle during the early 80's recession.

California had diversified way beyond military contracting by the time the 1980s rolled around. I'm reading Kevin Starr's wonderful history of California during the big growth phase ('Golden Dreams: California in an Age of Abundance') 1940-1965 and things began to diversify in the late 1950s. The recession in the beginning of the 1960s was far worse for the state than the one in 1980 because the dependence on military work. I had a first hand view of this growing up during that earlier period and my dad's architectural firm had three years of zero net income from 1962-64 because of the loss of the aircraft industry during that period. General Dynamics was gradually shutting down the Convair division and their attempt to diversify into civilian aviation was a total failure.

Orange, In 2012 it is still high, so in 1980 it was even higher.

Here is something recent from the Santa Cruz newspaper:

Tens of thousands of California jobs are at stake as the Pentagon rolls out plans to reduce its budget for the first time since the 1990s.

Though California is far less dependent on defense dollars than it was two decades ago when drastic cuts cost the state nearly half a million jobs, communities from the Bay Area to San Diego will feel the pinch.

Nearly 126,000 California jobs could be affected if defense spending shrinks $1 trillion as anticipated by the Pentagon, according to an economic analysis by the Aerospace Industries Association.

The Defense Department recently announced plans to trim spending by $487 billion over the next ten years. In addition, a stalemate over how to reduce the debt will trigger up to an additional $600 billion in cuts unless Congress finds a way to circumvent procedures it agreed to last summer.

Isn't China also a good example of Keynesian success? The 2009 stimulus prevented any significant Chinese slowdown.

I am confused like Armand - China completely avoided recession during the global financial crisis despite a huge drop in exports. GDP growth (annualized) never fell below 6% in any quarter, and once stimulus was really cooking, China posted 9-10% annualized growth rates. Even with the current slowdown worries, we're talking about falling to 5-6%. If this is truly a Keynesian story, then we should all be Keynesians.

If they ended up beating their long-term growth path, their stimulus was probably too effective and too large, even (hence a feeling of waste...?).

Or is 5-6% a disaster in a country that requires 9-10% to not be a disaster?

What is the baseline here?

Maybe we could trade places with them and get 5-6% repressing our people for a few decades beforehand. We didn't plan ahead!!!


With tens of millions of people per year exiting rural poverty to join China's more modern economy, a large amount of growth is required just to tread water.

And Matt Yglesias makes the point that China is a success over the last few years because they have growth that could only be possible by exploiting catch-up growth potential. Keynesians are insane, or at least I'm missing something more subtle in their text.

BTW, people don't read carefully, but I don't blame them. TC expresses no opinions here.

I simply ask "what is the cost?" and every time they say "free lunch" my impulse is to say "go back and look again."

Not taking sides in this Keynesian debate, but just how is helping tens of millions of people per year exit rural poverty to join China's more modern economy "treading water"?

I'd call that progress to be trumpeted from the rooftops, no?

Because they should have been there already!

So should we, but like China, we saw fit to nurture ourselves a large crop of million- and billion-aires instead.

@ dead

Because the hundreds of millions who have already left poverty want to see growth in "their economy" as well.

But wasn't the argument in the US that a large part of the stimulus was to prevent regular government jobs from being cut due to state budget problems rather than technocrats creating new programs? Also isn't the Keynesian position more about timing of government investments - frontloading future spending rather than a position about how much government should be spending in the first place? A Keynesian economist could simultaneously be very critical of the level of state expenditure in China, and still say that if you are anyway going to move on that trajectory, it would be good to shift investments to a recession year.

Good point. But there is no theoretical reason that government spending has the same returns (net present value) as private investment. And it's not necessary. Even if you assume a very ineffective effect of government investment, enough spending will get you results in current period transactions. The issue is not about the nominal measure - it is about the real effects of real goods and capital allocation in the economy and their efficacy to promote current and future consumption. Just as K is not uniform, investment opportunities are not uniform between periods of time.

I third Armand. I would take crap bridges and full employment over rotting roads and masses of long term unemployment. And if we could use our superior institutions such as rule of law and contracting regulations to get something useful out of the deal, so much the better

Yes, I too would like some clarification of some of the points made above. From what I've read, China has been hailed as a Keynesian success story. Its current problems also don't seem directly related to the stimulus, but with broader systematic problems.

Collapsing bridges, collapsing solar and steel prices, a collapsing real estate market. Insolvent banks and insolvent local governments. It's worse than 2008.

Have we gotten so old that a new generation who doesn't remember 2008 has already risen?

Are you talking about the US, or China? Tough to tell.

In any event, the collapsing real estate market is a private (read: nothing to do with government) phenomenon and thus cannot be blamed on Keynesian spending. Re: collapsing bridges, the explanations I've read indicate that they were built on the cheap, thus lending credence to the opposite argument: perhaps too little has been spent.

I wouldn't say NOTHING to do with the government.

I'm not getting the issue with China's collapsing bridges. Are you guys saying the bridges weren't needed? Then no one was using them so who cares if they collapse? Saves everyone the trouble of tearing them down. If they were being used by lots of people then they are needed and that points to the need to have spent the money necessary to build them correctly.

Great Post. I think the debate above is more to do with the difference between an increase in output vs a sustainable increase in output. A sustainable increase in output requires that the investment generates a return on capital - something that Keynes was clear about in the General Theory with regards to any stimulus programme (and that nearly all neo-Keynesians have subsequently ignored). This also explains why of all the BRICs, China's equity market has performed so badly despite its consistent higher level of output compared to say Brazil. During the lost decade in Japan, several regional airports were constructed (paid for with increasing government debt). Many of these airports are in effect loss making assets. This is not what Keynes had in mind. Japan's debt to GDP ratio is 230%.

You are misinterpreting Keynes. The main concern of Keynes was not "sustainable" increases in output initiated by government action. Rather, his concern was that of returning to full employment as quickly as possible following a fall in output and of moderating the fall in output when it happens. On that point, here is Keynes in Chapter 10:

If the Treasury were to fill old bottles with banknotes, bury them at suitable depths in disused coalmines which are then filled up to the surface with town rubbish, and leave it to private enterprise on well-tried principles of laissez-faire to dig the notes up again (the right to do so being obtained, of course, by tendering for leases of the note-bearing territory), there need be no more unemployment and, with the help of the repercussions, the real income of the community, and its capital wealth also, would probably become a good deal greater than it actually is. It would, indeed, be more sensible to build houses and the like; but if there are political and practical difficulties in the way of this, the above would be better than nothing.

Sometimes I wish the distinction between monetary and fiscal policy were clearer in the GT.

Ricardo, there are two areas you might want take a further look at in the General Theory. Firstly, Keynes' quip was clearly akin to mining which is a value added activity. However Keynes does not assess how much it might cost to actually get treasury officials to set the project up. For example the security operation would be extensive and it also assumes that the labour needed to extract the money would be available. However, mining is a skilled activity due to the capital intensive nature of it so on several counts this example is not particularly well thought through in the book. Perhaps in Keynes' mind, the army of unemployed workers would turn up only armed with shovels and the mining process would be akin to an 18th century Welsh coal mine.

The second part is why Keynes was at pains to assert the importance of a positive return on capital is precisely because of his desire to bring about full employment as you say. An injection of an investment that had no multiplier and was not able to maintain the employment would not bring about full employment. See Notes on the Trade cycle at the end of the General Theory.

For example the security operation would be extensive and it also assumes that the labour needed to extract the money would be available. However, mining is a skilled activity due to the capital intensive nature of it so on several counts this example is not particularly well thought through in the book

Or perhaps you are overthinking it. Clearly the labor and capital is available to bury the banknotes so it would be available to dig them back up again.

But granted, stimulus that's set up to be impossible to get is functionally the same as no stimulus. If the Treasury printed twenty $100 Billion bills and offered it to the first person to build a faster than light space ship they might as well not bother.

Another +1 for Armand. The point of Keynesian stimulus is to stimulate aggregate demand and prevent a collapse to a suboptimal low-work low-consumption equilibrium, not to improve long-term economic efficiency. I think everyone agrees (as Keynes did) that if you're going to do Keynesian stimulus you should try to spend it well, but the whole point of Keynes's digging-and-filling-holes example was that the efficiency of the projects being built through stimulus isn't critical to the overall argument (if nonetheless important in welfare terms).

To all of you who think this is a rather successful Keynesian story: if all it did is maintain output for a year or two, at the price of considerable debt, and a lot of corruption/inefficiency, and it did not change things in the long run (I would say it made them worse because more debt money was poured in construction), then really, what good is a stimulus anyway?

Do you know what keynes meant in the long run we're all dead?

Yes, what if the "bridge" just gets you further out on the edge of the cliff.

Time for Keanu Reeves "I'm going to speed up!!!"

>Do you know what Keynes meant in the long run we're all dead?

He meant, my policies are not sustainable over the long term but what the hell, we'll all be dead by then and it's somebody else's problem.

Keynes was a member of a, ahem, group not generally associated with long-term thinking.

And he was in a more serious time when not getting through the short-run literally could end up in millions perishing. Nowadays it could result in the end of humankind, which is unthinkable, so we don't think about it.


You're being unrealistic here. Corruption in China's gov't was almost certainly not created or even enlarged to any real degree by it's stimulus program. China's gov't is corrupt because, surprise, it's a one party Communist state that refuses to recognize free speech or a free press or submit it's leadership to democracy.

Likewise China's rapid economic growth and resistance to the recession shows that it's not increasing in inefficiency. It is getting more efficient but it's starting from a very low base.

As for 'making things worse' because of debt.

1. It's highly unlikely that every bridge and road built in China is unuseable.

2. All major capital projects are build 'by debt'. It's the most rational way to build such projects. Huge capital projects provide benefits over extended periods of time. Debt means you pay off their cost over time. Consider the Lincoln tunnel which was built about 100 years ago and still provides a route for tens of thousands of cars every day. Even after the sons of the men who built it have long gone to their graves, the project is still providing us with great economic benefits today. The matching principle makes perfect sense that you pare a project with benefits over time to payments over time.

3. In terms of debt problems, the shining examples of austerity prove that you are chasing fools gold. How many times have we now seen a gov't promise to 'control its debt' with austerity in the face of a Keynesian recession only to discover six months after passing cuts in spending and tax increases their deficits are now higher because increasing unemployment has driven down tax collections and increased welfare payments despite the cuts?

3.1 I'll demonstrate to you another way debt from stimulus is no problem. Let me ask any serious person who worries about the US fiscal debt to address this hypothetical:

A UFO lands and the alien comes out and announces that as a gift to Earth from the Cosmic Alliance, he will grant one of two wishes. He will either produce a way to bring Medicare and Medicaid into long term fiscal balance forever. Or he will write a check to cover the amount of US debt that is represented by Obama's stimulus (about $750B let's say). Would anyone here opt for the second? If not then doesn't that demonstrate how utterly trivial 'debt from stimulus' really is?

Furthermore, if anything, China's gov't has actually gotten less corrupt since the stimulus.

"doesn’t that demonstrate how utterly trivial ‘debt from stimulus’ really is?"

No! It demonstrates how insane the entitlements are without restructuring!

It's not like "hey, we can fix the estate tax...or not and just remain a joke." If we don't fix is no if. They absolutely will be fixed even if it simply default of some form.

Do you have an inside track with these aliens?

"A UFO lands and the alien comes out and announces that as a gift to Earth from the Cosmic Alliance, he will grant one of two wishes. He will either produce a way to bring Medicare and Medicaid into long term fiscal balance forever. Or he will write a check to cover the amount of US debt that is represented by Obama’s stimulus (about $750B let’s say). Would anyone here opt for the second?"

"Do you have an inside track with these aliens?"

I'm pretty sure I know what their fix for Medicare is:
To Serve Man -

proves my point, the 'debt' of the stimulus is trivial even in your book. And as you say if we don't 'fix' entitlements the debt issue swamps everything else. So in other words carping about the 'debt' caused by the stimulus is totally irrelevant. If we never incurred the stimulus debt but left entitlements heading on their track everything's going to blow up anyway.

Therefore why focus on the debt of the stimulus?


1. So what if they are usable? Investments have to be justified on their ability to pay for themselves, not on the fact that they sound like a good idea. Ferrari's for everyone would be great: everybody getting very fast to where they're going, in style. That doesn't mean we should purchase such cars through a massive stimulus.
2. Building by debt is fine. But again, as per #1, who pays for the debt, and the risk? Entrepreneurs in a free(er) market, free from government planning are under incentive to build profitable projects. Agents of government act under entirely different incentives. And then you have crowding out.
3. I put it to you that no government in the world has attempted serious austerity. Talking about is does not mean implementation. Social democracy being what it is, it is simply impossible, in my opinion.
What you talk about is continuing the party because confronting the morning and the inevitable hangover is to hard. If the whole economy has been put on a wrong track, one way or another, that has to be addressed. Again, you're thinking about short term consequences, and ignore the long future.
3.1 " If not then doesn’t that demonstrate how utterly trivial ‘debt from stimulus’ really is?"
I ask you, for how long can you maintain that question? 100% of gdp? 200% of gdp? What if this can be made to work for hundreds of years, the state amassing debt for generations? Still no problem?

I feel you have not addressed my main point because you cannot conceive that anything is more important than just keeping it going, at all costs. Full employment, gdp growing , even if everybody is digging and filling holes, and there is a mountain of debt to be collected.

I feel you have not addressed my main point because you cannot conceive that anything is more important than just keeping it going, at all costs. Full employment, gdp growing , even if everybody is digging and filling holes, and there is a mountain of debt to be collected.

Let me take a stab at it again then.

I think your are assuming here the debt has a real impact just as the unemployed do. I think the debt issue is only nominal.

We know there's a real cost to unemployment in a Keynesian recession. Perfectly capable workers sit idle rather than producing output...output that raises GDP and can be used to pay down debt, invest, enjoy and so on.

So in order for you to establish that restoring full employment via debt spending is a net bad, you have to demonstrate that the increased debt, at some point, will cost us more than the output produced by restoring full employment months, years, earlier than it would if we just let everything be.

So tell me what mechanism you think this might be? The only one I can think of might be that those who lent us the money want it back. OK, say China wants its debt paid off rather than buying more. That begs the question of what China will do with the cash? Buy US goods? Bye bye US unemployment. Open a savings account at some little bank in the midwest? Well that lucky bank is either going to make lots of loans to US consumers and businesses (bye bye unemployment again) or buy US Treasury debt (back to the same place).

The only problem I think that might happen is those who loaned us the money get their cash back, and want to buy US goods which is more than we can produce even if we put everyone to work thereby sparking inflation.

BUT, here's a problem. The most we ever borrowed for stimulus was out 'output gap' to full employment or less. Unless the US suffers some major supply side problem (say that UFO deciding to nuke the entire West Coast), there should be no problem with a full employment economy producing sufficient goods and services to trade for those dollars.

Do I get to take a point away from Armand if I think Keynes was a fraud?

The point of Keynesianism is Keynes. His ideas are idiotic on their face, per the example provided by Ricardo which is nothing more then the "broken window" fallacy rewritten, but do perform one function quite well - provide academic cover for politicians to do what they're viscerally inclined to do, spend other people's money.

Keynesian silence on China is explicable by caution in the face of the most recent failure of the great man's ideas here in the U.S. Keynesians have their hands full explaining why the most recent Keynesian stimulous failed in the U.S., their explanations meeting laughter and sneers where they're not greeted by an upraised digit in response, so far better to remain quite on this other example of the application of the great man's ideas until it's determined whether, in a departure from the previous, uniform history of failure, it's a success.

Much easier on the great man's legacy to quietly ignore another failure then to have to explain it.

If someone wittingly rewrites the glazier's parable but stresses a counter-intuitive resolution of the allegory, then complaining that they've completely missed Bastiat's boringly obvious conclusion probably you've entirely missed the point. Do you think Keynes is unaware that burying money is wasteful?

probably *means* you've

I think Keynes' point is that the point of money is to get people to do stuff. Burying money might be the most unwasteful stimulus in a purely Keynesian recession. It doesn't waste anything else, occupies people, and buried money is, ironically, sterilized.

That's pretty much the opposite of Keynes' point. As he says in the passage I quoted above, "It would, indeed, be more sensible to build houses and the like; but if there are political and practical difficulties in the way of this, [burying money and digging it up again] would be better than nothing."

Ironically, I suspect that is because he would assume that more houses would always be useful. Not so much, it seems, as we monthly get new proposals to bulldoze houses. What if he knew that all spending would be "wasted." What would be the least wasteful? Something that doesn't use up natural resources at the same time we are wasting...I mean...occupying (it's a feature, not a bug!) man-hours.

Houses, oddly, don't have long lasting value. If they are not occupied by someone and an owner isn't taking care of the house it falls apart pretty quickly to become so unusable after a few years as to make it easier to bulldoze it and start again. Also it's not easy to move houses around so if an entire community is on the outs, you're going to see a lot of houses having to be bulldozed.

Yet if you look at the number of houses to people it did not increase all that much during the 'boom'. The housing bubble was a bubble in housing finance, not houses. Yes stupid houses were build but we are hardly awash in a glut of 'too many houses' the way we had a glut of 'too much fiber optic cable' after the dot com boom.

Is unemployment insurance more efficient? Are bank bailouts more efficient? Last time I checked, the debt levels of developed countries rose more than those of China, and indeed the debt levels of China have stabilized since the crisis, while those of the developed world have not. China avoided recession, and invested in a lot of things that have an uncertain but potentially large social return in the medium term. Solar, for example. A disaster now, but I would not be surprised if ten years from now this investment is looked upon as the defining factor that put China at the frontier of this technology. Social returns are different from private returns, and the Chinese are really good at identifying that. I think the US used to be really good too. Would there have been a space program if people were focused on the "ROE"? What was the ROE of the interstate highway program? What is the payback period, and to whom does it pay back?

Debt levels of developed governments went up. U.S. total credit has been mostly flat, with U.S. government debt offsetting private deleveraging. China didn't do a swap of private for public debt, they increased total credit on a massive scale. The costs will be born by the Chinese people via financial repression, just as they paid for the excess of the 1990s during the 2000s.

If this "paying for excesses of the 90s," sign me up:

One wonders if large scale private sector investment is ever cronyistic, corrupt, and productive of malinvestment. There fact that money has been spent in corrupt ways and that there are unproductive projects and "diverted" resources seems entirely uninteresting. The existence of such is enough to damn the market as well. So how do things compare?

It only matters if I am forced to pay.

Poor baby, paying taxes. It must be horrible. I'm amazed you don't just shoot yourself due to the agony of it all.

The question hinges on to what degree was China ever in a Keynesian recession (aka balance sheet recession) to begin with? In a non-Keynesian recession the world 'looks normal'. China's stimulus then would be little more than an increase in infrastructure spending, which may or may not have been a wise investment.

+100 for actually displaying nuance.

Far too rare in any Kenesian discussion.


Honest question, where do you find that a "balance sheet recession" is a "Keynesian recession"?

I would say a Keynesian recession is one where demand for actual purchases falls dramatically. A balance sheet recession is one where people or businesses suddenly decide they have too much debt. An asset bubble is highly likely to cause this since during such a bubble a lot of people end up taking on debt to buy assets that suddenly become worth less. Just the sort of thing that would make you want to pull back your spending and try to get rid of your debt.

When everyone at the same time, though, decides to pull back spending to reduce debt incomes have to fall. To use Paul Krugman's way of putting it, consider that all spending can be grouped into three categories: consumer, business and government. What happens if you are a, say, a landscaper and consumers decide to pull back on spending? You may try to get business contracts. What happens if business pulls back? You'll try to get gov't work. What happens if gov't either pulls back or doesn't increase enough to offset the pull back of the other two sectors? You're going to be in trouble and if you're doing landscaping because you're trying to pay down your debts you're in double trouble. But note there's no 'supply side' problem there. You didn't buy crappy lawnmowers or weed whackers. Your truck works perfectly. You are perfectly capable of producing lots of landscaping services.

I agree with you that you can have a Keynesian recession without a balance sheet one. Anything that causes a massive pull back of total spending would work. I suppose if one company employed a huge portion of the population and it suddenly came to light that their entire pension fund was lost in a ponzi scheme both business and consumers might also pull back spending dramatically....even though debt isn't a problem. In a small country like Iran, if its export market suddenly collapsed (say by massive sanctions imposed on it by the outside world), that to might do it. But for a large country a lot of those things tend to cancel each other out....except for bubbles which can feed on themselves and grow extra large in a large economy.

I would have thought that people in the developed world would be looking to Australia as the prime example of Keynesian stimulus in the 21st century. When the extent of the GFC started to become clear, Australia crafted a stimulus package calculated to be large enough to make up for the expected economic downturn, while keeping interest rates high enough so that they could be dropped if it became necessary to quickly respond to unexpected shocks. As a result, there was a slowdown in Australia, but no recession, and it has since been by far the best performing economy in the developed world.

And because people people always seem to bring it up, I will mention that Australia did not avoid a recession as a result of its mining sector, mining suffered a greater slowdown as a result of the GFC than the rest of the economy.

GFC = Great Finanicial Crises?

Per the theory, stimulus helps offset a drop in demand, it does not help a contraction of supply. I'd like to know more about what happened to Australia's mining sector but if a country makes a lot of its income from a mine and the mine dries up....or the rest of the world stops buying the mine's output for the same prices it used too in the past...that would cause what counts as full employment in that country to fall. Stimulus would not help with that.

That is why Republican orientated economists have been putting heroic efforts into trying to find some supply side centered cause of the Great Recession (i.e. a complicated tax code, regulation, and my favoite 'uncertainity' which too often sounds like a polite way of saying "business is just too freaked out by a President named 'Obama' to invest). They haven't been too successful since they haven't really come up with anything that wasn't a fact of life long before 2008 (the tax code was pretty complicated under Reagan, Bush, Clinton and Bush II, so was regulation, so was 'uncertainty' when no one knew how the first Gulf War would turn out, no one knew what Clinton's health proposal would be or if it would be enacted and so on). You can always find proposals to improve our supply-side, but that's not the same as saying you found the cause of the Great Recession or its solution.

But the world is a big place and I can see how smaller countries can get whacked with huge supply side hits. For example, if tomorrow some crazy scientist invents a 'Mr Fusion' to power cars on kitchen garbage, Saudi Arabia is going to suffer a major supply side hit and stimulus won't save it.

Well maybe if Delorean had a bailout we'd already have these magic cars!

Don't worry, Teslas for all!

What is the baseline of corruption or bad projects with regard to Chinese infrastructure spending? China is an enormous rapidly growing country and it is also a country that stimulus or no is considerably more corrupt than the west. I think it would require a lot of knowledge of Chinese infrastructure projects over a medium term time frame to even begin to evaluate how this stimulus effected the quantity/quality tradeoff some are describing. Inquiry by anecdote on this topic seems to be nothing more than opportunity to reason by confirmation bias.

Only if one is what I might call a knee-jerk conclusionist. Here is the timeline.

TC posts a question. I say to myself "Hmmm." The other side tears their shirts off and says "resounding success!" Then I have to say, "hang on a second!"

Your first few sentences are great, and to just ask the question doesn't negate all that reasonableness as in your last sentence until everyone starts tearing their shirts off.

In practice, large-scale government stimulus is an invitation for corruption and a diversion of resources that builds up knowledge and capital in unproductive areas. As Obama said after the US stimulus package: there’s “no such thing as shovel ready projects.” Even in China.

I don't think the evidence proves this out. This represents a classic case of where a textbook illustration has overtaken actually looking at reality.

First the illustration: Stimulus = picture of construction worker building a road/bridge/tank/etc. Good action orientated picture to keep students interested in reading an economics textbook. But is it really reality?


1. If you went to, say, NASA, and told them that there was funds equal to about 5-10% of their yearly budget that was available for an additional project, provided they could get it done in the next year or two, what exactly do you think would happen? Would they waste it on covering the Space Station with gold leaf lettering saying 'NASA Rulz BoyZs!!!'? Probably not. They'd probably do an additional Martian probe, maybe test a few new things or what not. Same thing if you did that exercise with the DOT. Any large organization has in its budget a list of things that almost but didn't make the cut. You have to keep a list of those things because sometimes money becomes available and if you can ask for it quickly enough you can get some of your 'wish list' items accomplished.

2. More importantly, very little of stimulus had anything to do with shovels. The bulk of stimulus as it was done in the real world was increases to food stamps, increases to unemployment, cutting payroll taxes to increase takehome pay and helping states fund Medicaid to keep state taxes low(er) and avoid (for a time) massive layoffs on the state level. When you back that out, you're probably talking at least 70% of the stimulus as it was done in the real world. What's left is a bit of 'shovel projects' but this, unfortunately, gets all too much attention because it's what people think of as stimulus because they remember the pictures in the textbook mroe than the text.

2.1 what's interesting is that these examples are essentially 'nonbiased'. You can claim the bailout of GM had to do with the jobs it provides in a Democratic swing state, the willingness of its execs to support Democrats or the benefit of helping a unionized company. Granted, but things like unemployment insurance are essentially unbiased. If you loose your job you get it regardless of how Red or Blue your state is and it's not more tempting to loose your job and collect than keep your job. Most of stimulus is no more biased than montary stimulus (i.e. lower interest rates). When the Fed lowers interest rates, it doesn't decide who gets a car loan or who gets a small business loan. Banks decide and even with lower rates if they decide wrongly they will suffer financially for it.

"More importantly, very little of stimulus had anything to do with shovels." This is an important point when we think of how stimulating all these infrastructure projects are. The world has changed, construction work no longer involves thousands of workers doing manual labor, it is a specialized task that is performed by a small amount of skilled labor and has little to no effect on the average unemployed guy.

Granted Mofo but stimulus's impact is on aggregate demand. Always has been. Even during the Great Depression very few people out of the entire population worked on the Hoover Dam or for the WPA (in fact consider even when unemployment was 25% that meant 75% still had a regular job!)

I don't think that is the reality. The Obama quote is a real quote. I think the politicians went to design a stimulus plan and found that there weren't a lot of projects with positive ROIs that were ready for investment. That was for a stimulus that many argued was too small.

In contrast, the stimulus in China was very large. A Keynesian dream, right? As a result, banks got the directive to lend huge sums of money and SOEs got the directive to put it to work. Where? This was not a 5-10% budget increase so it was not easy to find places to spend it but generally it found its way into real estate, infrastructure, steel, and other similar projects.

Here is a relevant article from today:
"After the financial crisis, when the government released its stimulus, banks begged us to borrow money we didn't need," Li Huanhan, the owner of Shanghai Shunze Steel Trading, told a judge at a recent hearing. "We had nothing to do with the money, so we turned to other investments, like real estate."
While some loans did go towards equipment and expansion, executives admit money was also used for pet property projects, plush apartments and stock market bets.
By the end of last year, China's steel industry had a total debt burden of $400 billion - around the size of South Africa's economy. Some of China's leading mills alone owe 200-300 billion yuan ($32-$47 billion), according to the China Iron and Steel Association.

The science-funding points are the most interesting to me. The after-the-fact criticism that we didn't have enough shovel-ready projects, as it turned out, isn't right at all from my limited perspective. We could've easily spent at least tens of billions more in additional science research stimulus, where the main things we spend money on are employment and modern technology (much of which is American-made), both of which, regardless of the impact of the actual scientific research being funded, have great long-term benefits for the economy in addition to being good short-term stimulus.

Yggy's site requires registering to comment, so I'll post this here.

"But look at the numbers. In 2007, China's PPP-adjusted GDP per capita was $????. In 2008 it was $????. In 2009 it was $????. In 2010 it was $????. In 2011 it was $????."

There, I fixed that for you Matt.

If the Chinese government spends a billion dollars on something that provides no value to anyone, it is a fallacy to count it as a billion dollars of GDP. It is zero dollars of GDP. And yes, a ghost town with zero inhabitants provides no value to anyone. I'm sure the construction crews earned income, but it would have been more efficient to just write them checks and not waste the real resources.

And yes, the US has this problem too, as does every other country that engages in wanton government spending. China just seems to be making a career out of it rather than a hobby.

If the Chinese government spends a billion dollars on something that provides no value to anyone, it is a fallacy to count it as a billion dollars of GDP. It is zero dollars of GDP.

The Koch brothers are spending a billion dollars to build a wild west themed ghost town. Should we back that out of GDP calculations for 2012? Is the Mars lander of no value? Should we back that out of GDP but leave in the occupation of Afghanistan? How about all the money spent on Lady Gaga songs?

Thanks but no thanks. GDP accounting is best left alone rather than letting people like you substitute in subjective guesses about what is valuable and what isn't.

This is an amazing statement from ygs. It seems literally impossible to argue against government spending, regardless of the type of recession or any other variable. All you have to do is put one dude to work.

"But there's no sector of the economy with lower productivity than the unemployment sector, and no public investment more wasteful than having a human being spend years scanning help wanted ads and feeling depressed. Does someone seriously want to argue that none of these projects have any value?"

It seems literally impossible to argue against government spending, regardless of the type of recession or any other variable.

It seems like you assume recession is the norm, it's not. When the economy is at full employment (and no that doesn't mean zero percent unemployment) it's quite easy to argue against gov't spending. Likewise it's quite easy to argue against gov't spending for recessions that are purely supply side in nature.

Repost due to some kind of connection error. Apologies if it duplicates.

Why would the specific argument being put forth by Yglesias here be constrained by a recession condition? If the government can put one guy to work and the return on investment is completely irrelevant ... doesn't that many any spending for any employment gain is fine?

Ok so you're backing up a bit. You're saying Yglesias' argument is about recessions, before you said it was about gov't spending any time anywhere.

I think you're equating recession with unemployment. But technically recession is a steady decline in GDP. You don't need to have unemployment for that to happen. Consider a possible supply side recession. Saudi Arabia makes most of its GDP by drilling and selling oil. One day they invent 'Mr Fusion' and suddenly all cars cease burning gas. Saudi Arabia's GDP will drop like a rock and there may be a brief peirod of intense unemployment but over the long run it will get back to full employment, but at a much, much lower income level than before.

Perhaps one could argue that some type of gov't spending could transform Saudi Arabia's economy away from oil towards something else. But that's less of a Keynesian stimulus argument and more of a central planning argument.

Stimulus in itself won't stop that. The US had something like that happen to it after WWII. There was a sudden surge in unemployment as hundreds of thousands of GIs returned home from the war, but in months that disappeared as the domestic job market expanded to accomodate them.

No, I get all that. I just am struck that Yglesias specific argument in the part I quoted suggests that it should always be true for all unemployment at all times that it's better to spend any arbitrary amount of federal dollars as long as one dude is put to work in the process.

Or to put it more bluntly, in a supply side recession, the unemployed are part of the market forces at work. Whether it's Saudi Oil workers in after the invention of some magic juice to replace oil or US GI's returning home after WWII, the unemployed are the market economy finding where to put people to work.

In a demand side recession the unemployed aren't part of the economy reallocating itself, it's just the economy making a perfectly useful resource idle.

Hence the almost sad efforts by Republican orientated to paint this as a supply side recession (with the 'sins' being basically 'electing a Democrat'...even though the recession started and most of its damage was done after quite a few years of Republican rule)

But do you really believe the U.S. has a shortage of domestic demand? We are still in current account deficit, after all. President Obama himself believes this is a supply-side problem (as famously reported in Confidence Men), much to the horror of his advisers.

America's problems are things like lack of external demand, nominal divergence of regional price levels (San Francisco and Utah have similar median real wages, for example, but wildly divergent nominal wages), low and uneven productivity growth, etc.

A supply side recession would result in high long term interest rates and inflation in response to the efforts to stimulate the economy.

America’s problems are things like lack of external demand,

That's demand, not supply. Even if demand is coming from the rest of the world, it's still demand.

nominal divergence of regional price levels (San Francisco and Utah have similar median real wages, for example, but wildly divergent nominal wages)

Why would that be a supply problem? You are saying a plumber in San Francisco gets paid more than a plumber in Utah, but the San Francisco plumber finds prices for things like housing, food, entertainment etc higher than his Utah counterpart. In what way does that inhibit the supply of either plumber?

, low and uneven productivity growth, etc.

Negative productivity growth may, I suppose, generate some type of supply side recession. If tomorrow half of us became stupid(ier) I suppose our GDP would fall. But zero or low productivity growth would not generate a recession, only lackluster economic growth. 'Uneven' growth would be a positive for GDP because just as the world gains from free trade, individual nations gain from internal free trade.

Let me refer to the parable of the world's greatest lawyer whose also the world's fastest typist. Since he earns much more for billable hours, it still makes sense for him to hire an average typist rather than doing his own typing.

As for whether or not Obama believes this is a supply side recession. Why do you think that's relevant to this discussion? His proposed policies have mostly been demand side. Granted some policies like improving the educational system may be deemed supply side but those are sensible policies even if you don't think we are in a supply side recession. Improving the supply side, after all, is a positive for economic growth.

More problematically for you, if you think this is a supply side recession, you have to tell us exactly what happened in 2008 that has been reversing very slowly since? Are you going to tell us that nominal wages in Utah and San Francisco were even during the 1990's and early 2000's? That productivity growth was exactly even across the US?

Sorry, I'm too used to lumping the current slow recovery and the recession together. Slow growth in the United States is a supply-side problem (and yes, I do think regional divergence is a problem for growth; at the very least, it makes labor mobility less effective).

Actually negative GDP growth is another matter. I agree with you there, though personally I prefer discretionary monetary policy and non-discretionary fiscal policy (automatic stabilizers).

Also, I'm not sure I understand your point about demand. The U.S. does have an overabundance of internal demand. That's inarguable, given our low savings rate and persistent current account deficit. What we lack is external demand.


I'm just using very simple supply demand curves. Where they intersect you get price on the y axis and quantity on the x axis. Now we have a recession which means quantity is low. You claim this is due to our supply curve being low, or shifted to the left. Well do the sketch and what do you see. Shift a supply curve to the left and you'll see quantity fall but prices rise. That is not what we see in the US or world today.

You claim the US has a low savings rate, but to cause a supply problem this would mean it must be difficult for businesses to get loans for capital investment because savers are supply few funds. If this was the case we'd see high interest rates yet we see low interest rates. We also see the corporate sector amassing huge stockpiles of cash balances. (and no don't tell me it's the Fed doing it, the Fed only controls the 3 month rate. If the market thought there was a shortage of savings or inflation was looming they'd send long term rates skyrocketing).

I hardly think it's inarguable that the US has an overabundance of internal demand. Again look to your simple supply and demand curves. The supply curve slops upwards until it reaches our max. possible output at which point it becomes vertical (once you've got everyone working on producing something, more demand can't get more of the product produced because there's no one left to put to work on it, you can only raise its price). If we had surging demand we'd see reduced unemployment and inflation heating up.

Likewise there's no difference between 'internal demand' and 'external demand' here. Demand for goods and services is demand for goods and services. Business isn't going to refuse to hire employees to meet the demands of paying customers because 'too much' demand is coming from 'internal customers'.

Boonton, your insistence that foreign demands "doesn't matter" is naive. As it happens, businesses do hire based on where the demand is coming from.

What happens to the dollar--and unit labor costs--when consumption skyrockets, and savings falls? The dollar and labor costs both rise, and we get a current account deficit. Now, you claim that internal demand has no effect on jobs. But if the cost of labor is artificially high, firms can (and do) create jobs overseas.

As for inflation, the Asian savings surplus (and associated deflationary pressure) has eased inflationary pressures in developed countries. After all, if we don't have enough supply here, the Chinese are more than willing to increase supply there. The American economy is not autarkic.

Finally, we do have a savings problem. Most saving is either done by corporations, or more rarely the federal government (as in the late 90s). This is not healthy. If American consumers saved more their wages wouldn't need to fall to achieve current account sustainability (at least, wages wouldn't fall as much).


What exactly is wrong with that statement?

Given a choice between an economy A where ten percent of the population is unemployed and doing nothing, and economy B where all ten percent of those are working but only 2% are engaged in productive work, why would we not always, always choose B? Especially given that there is abundant psychological evidence to indicate that the mere status of being "unemployed" has psychological costs that translate into economic costs.

Because, as the depression, Japan and numerous other examples show, the situation becomes perpetual. You get the nasty situation where youth unemployment gets cemented between 25-50% and continuous stagnant growth. Spurts of activity happen coinciding with periods of government largesse which don't last due to fiscal or political pressure. The business class loses its ability to compete in the real world, becoming adept at manipulating government for income instead of building markets.

And then the money runs out. It doesn't work.

Would it be possible for the 8% being employed on "unproductive work" (to get all 10% working) to actually be a drag on the economy and theoretically end up with a net loss to the economy?

I have my doubts that psychological effects of doing completely pointless tasks are all that different from those of being unemployed. It's kind of Kafkaesque to put people to work building empty cities. Chanos is not a "Macro" investor but he does a good job of explaining the problems in the Chinese economy. Also Hugh Hendry

If only the defenders of the Keynesian type stimulus can spell out what type of evidence would persuade them stimulus does not work , and if the critics of stimulus can spell out what type of evidence would persuade them that stimulus does work, the two groups can have a productive discussion about the Chinese experience.

How about seeing the predictions of the anti-stimulus critics come true? Where's the onset of inflation? Where's the onset of higher long term interest rates due to massive gov't borrowing? I think your question should be turned around, how many times must stimulus critics see their theory's predictions fail to change their mind?

I submit this 5 minute talk from Patrick Chovanec on NPR this morning on declines in China's growth rate. Let's assume that China's growth is flat to declining right now. How does that affect the argument if we assume the stimulus spending temporarily kept the Chinese economy growing but didn't change the long-term growth in the economy?

This is absurd. China has a trade surplus. America has a trade deficit. In the natural order of things, China must consume more and America must consume less. Thus, Keynesian demand-side policies makes sense for the Chinese, who worry their export markets may collapse if they don't shrink their surplus (unlike Germany, the Chinese have learned that lesson).

America is much better served by loose monetary policy and tight fiscal policy. Not because monetary policy has a stimulative effect here, but because it can stimulate our trading partner's economies, appreciate their real exchange rates and create demand for our products on world markets. The U.S. has the unique ability to stimulate external demand by raising total global demand and total global income, as opposed to beggar-thy-neighbor devaluations. Monetary policy has accomplish this; fiscal policy can not.

"Monetary policy has accomplish this; fiscal policy can not"

"Has" should be "Can." I don't have the precise facts on the effect of Bernanke's policies on global income, and while I think they had an effect I can't say that as a fact.

When there is inadequate demand and the state does not have projects that yield a NPV at the discount rate, then the stimulus needs to come in the form of lower taxes and transfer payments (which will be appropriate to reduce distress of people whose incomes that will have fallen). It appears that the Chinese did not chose their projects will, but their GDP never dropped, either. As always when dicussing policy decisions, what is the opportunity cost?

Comments for this post are closed