Paul Krugman makes the argument for austerity

…it’s overwhelmingly likely that these [Social Security] projections will be hugely off in one or more ways.

No immediate moral from me, except as a reminder that all those long-term budget discussions in which people act as if we really know what the state of Social Security will be three decades from now are basically just boring science fiction.

Here is more, and this post too.  I bought, and still buy, that kind of argument when it comes to climate science and it seems to work for some number of fiscal issues as well.  Better safe than sorry.

Note, by the way, that to the extent the “robots” argument is true, some of the decline in work in the short-term will not be reversed anytime soon and thus the output gap is smaller — perhaps considerably smaller — than Keynesians have been claiming.


Climate Science, Austerity and Paul Krugman in one post? Does MR give a bonus for the year's most commented upon post?

And robots!

If there were zombies, it would set a record, easily. People like zombies.

Can we throw in Justin Bieber too? 1000+ comments no problem.

Lately there has two long term alarmist articles:

1) Robots will do everything so the labor demand will diminish.
2) Birth rates are dropping.

Seems like a long term economic evolution creative destruction to me.

And notice how both alarmist trends actually cancel each other out. Dean Baker has been all over this, just like he was over the housing bubble.

If there are no people, then the robots will have no work to do.

We're importing too many Japanese robots, they rend the social fabric and draw too many benefits from the social electrical nets.

Tenemos que importar los robots mexicanos, que hacen el trabajo que los robots norteamericanos no quieren hacer.

El mejor comentario del dia. Bien hecho!


Um, the billions and billions of us is expected to increase by billions and billions in the next decades. To be only a touch more precise, from 7 billion to 9 billion seems a reasonable demographic consensus.

Um, but p_a, you are correct we are headed for 9 billion. The problem is we are expected to top out around there and start declining. How does capitalism work when that happens? Looks like robots. Then the debate about how to spread resources among a greater and greater share of 'unproductive' people who no longer can work because robots do it cheaper and better.

"how does capitalism work when that happens?"

I don't see a problem with capitalism working in that environment. However, pyramid retirement schemes will run into problems.

But, then we will have a robot population problem. We will run out of oil to grease their joints.

"...all those long-term budget discussions in which people act as if we really know what the state of Social Security will be three decades from now are basically just boring science fiction."

But I thought Krugman liked the Foundation novels!

As opposed to Paul Krugman, Scientist! The guy's lack of self-awareness sometimes is breathtaking.

The robots argument reminds me somewhat of the immigration debate and jobs. At the critical juncture the debate becomes feelings.

In both cases, the idea is to be invited to give talks to people who think they are immune to losing their jobs so that the speaker can salve their collective consciousness. However, they needn't worry. They too will probably lose their jobs. They won't even be able to compete with the sex droids in the "new service economy".

So what's the _real_ answer?

Here's something he could have talked about but didn't because it is depressing to think about how hard it would be to implement given the powers that be:

A citizen's dividend paid equally to all adults that is financed by collecting the economic rent of civilization.

Since the powers that be have increasingly turned to rent-seeking, they will, of course, resent the very idea that "their" rents should belong to everyone. And they're in a position to make sure that doesn't change, aren't they? As a result, things _will_become dystopian.

This is why Japan could be a big winner. They didn't allow in any low skill immigration and their culturally homogenous society is more likely to share the fruits of the robot revolution.

Their population is cratering, though. Who's going to be left to enjoy those fruits?

Immigrants. Just like here.

There are a few Japanese still reproducing, they are doung something different from the Japanese that aren't and the children they have will reproduce. Barring some sci fi disaster, there will still be Japanese people having kids in a hundred years, and by then they will be past the demographic crisis.

This is why 8 may be right, while the population may have absorbed some SE Asian DNA in the process it will still be culturallt Japanese, and I always thought the Yamato race came from the South anyway.

Re: "Better safe than sorry."

Can I quote you when my Republican state legislators raid the "rainy day fund" one year to distribute money to taxpayers "because it is theirs", and two years later, when there is a deficit, there is no rainy day fund, and they won't raise taxes.

I will remember to quote you.

As to Social Security, didn't Alan Greenspan say that he was fixing Social Security would be good for 75 years with the changes they made in 1983?

True, but only if you willfully misread him. Otherwise, your comment is nonsense.

As "our leaders" make their 10 year budget projections, I was reminded today that 10 years ago the forecast was a 5 Trillion surplus, instead we recieved a 6 Trillion deficit. They only missed thier projections by $11 Tillion!

Right but that swing was mostly the result of a deliberate decision by the Republicans to bankrupt the country so they could squeeze entitlements for ideological reasons.

"squeeze entitlements for ideological reasons"? Um, Joe, I'm surprised you haven't got the memo yet, but Medicare was designed to blow up.

"Johnson maneuvered every step of the way getting this bill through Congress, and one of the things he did — and this is a little dicey in today's climate — was suppress the costs. So this young kid gets elected from Massachusetts, Ted Kennedy, in 1962, and Johnson is explaining to him [over the phone] how you get a health bill through. And what he tells him is don't let them get the costs projected too far out because it will scare other people:

"A health program yesterday runs $300 million, but the fools had to go to projecting it down the road five or six years, and when you project it the first year, it runs $900 million. Now I don't know whether I would approve $900 million second year or not. I might approve 450 or 500. But the first thing Dick Russell comes running in saying, 'My God, you've got a billion-dollar program for next year on health, therefore I'm against any of it now.' Do you follow me?""

Also, entitlements (Medicare Part D, Obamacare) haven't lost a lot of steam over the past decade.

From 4:17 to 4:47 is a full 30 minutes. I'm really accustomed to seeing an It Was The Republicans Fault post within 5 minutes. Please do better in the future.

And I'm accustomed to never seeing a Republican take responsibility for anything bad. If they ever start admitting that some of their policies have had, shall we say, less than optimum results, then maybe we'll see some better policies from Washington and local governments and maybe I'll even start voting for them again.

Don't see it happening any time soon.

Back in the "reality-based" world, Bush went out of his way to expand entitlements. Medicare Plan D is obvious. However, that's only the best known example. How about advertising (in Spanish and English) to promote food stamps? Ever rising "disability"? NCLB (entitlements for the education establishment)?

Bush never made any effort to constrain the welfare state. The Rove/Bush model was "goodies for everyone". Low taxes for the rich, Open Borders / Amnesty for Hispanics, endless war for Neocons, "free" trade / outsourcing for the Fortune 500, cheap labor for corporate exploiters, Medicare Plan D for the elderly, welfare for the poor, financial bubbles for Wall Street, housing bubbles for the middle clear, the "opportunity society" fantasy for immigrants, etc.

Of course, it didn't work and ended in disaster at home, abroad, and on the border. However, what would you expect from a something for nothing fantasy illusion?

What should be clear to anyone who is even remotely honest is that Bush never sought any restrictions on entitlements. Far from it. He sought to expand entitlements to protect his sacred cows (low taxes, "free" trade, Open Borders, etc.).

Yes, George Bush was the biggest liberal since Lyndon Johnson.

Still missing the point.

Doug M,

Joe Smith claimed that the Bush tax cuts were designed to squeeze entitlements. The exact reverse was/is much closer to the case. Entitlements were deliberately expanded to protect and enable the Bush tax cuts.

"Our Leaders" do not make 10 year budget projections. The CBO does based on current trajectories. How do you expect them to input political decisions into their models?

Economists can project growth and inflation reasonably accurately about 1 year out. Quality falls rapidly after that.

The CBO makes projections assuming that all else is equal. But all else is never equal.

Knowing that the 10 year number is BS, we should not calculate such a number. Anyone who believes it may have some basis in reality, will be led to make bad decisions. It is worse than wrong. We are even wrorse that the Soviets 5-year plans.

2 year budgeting!

It isn't an argument for austerity but for intervention though. The future is what we make of it.

Why do we need to import the building and construction robots from China? Why must we import the wind farm building robots from Japan. Why must we import the solar farm building robots from Taiwan?

Why can't America return to the capitalism that defined us before the 80s when capitalism became pillage and plunder of capital for short term profit with the long term future owned by the rest of the world?

After all,robots are capital assets and all capital assets are either natural and sustainable on the time frame of the asset (an oil field will be replaced in 500 million to a billion years),, or built by labor. You can not replicate a robot by download, but it requires labor at some point, unless people are less abundant than the robots.

"Why do we need to import the building and construction robots from China? Why must we import the wind farm building robots from Japan. Why must we import the solar farm building robots from Taiwan?"


"Free trade is Jesus Christ and Jesus Christ is free trade"

The dominant ideology of our time is cosmopolitan globalization. No matter the economic ruin it leaves in its wake, dissent is not allowed.

Free trade is sort of like Jesus Christ, except that it lifted hundreds of millions more people out of poverty.

Boom. +1 to Cliff.


This is some kind of public choice version of the law of gravity:

a. In good times: We have lots of extra tax money coming in. It would be irresponsible of us not to spend it/give it back in tax cuts. After all, this additional tax revenue clearly represents a permanent increase in wealth brought about by our wise leadership.

b. In bad times: Wait, tax revenues have gone down. How could this have happened? It would be irresponsible to cut any of the spending we raised during good times, or raise taxes on donors. Instead, we will do obviously dumb things, like furlough 10% of our teachers, policemen, and social workers, freeze cost-of-living increases in salaries, and underfund our pensions. But seriously, this is not our fault--who could possibly have known that the good times would end?

The first time I heard this song-and-dance, I even believed it.

i really dont see how your title follows from what you posted.

I can think of at least one person who would call this title a reading comprehension fail. Still it does place other angle on the posts. Let's suppose the CBO had put out a 'dystopian vision of the future' with rising inequality...would that honestly swing the balance now to fiscal austerity? I don't think the long term numbers are really the sticking point right now. The 'better safe than sorry' mantra seems to push toward the conservative (little c) forecast... reason why government agencies are wary to go down the dystopian path, neither the GTS+ or the cyborg revolution would pass as modal forecasts.

It might not be an argument _for_ austerity. But it sure is an argument for, we don't know what will happen. I don't know about you folks, but when I don't know what will happen, I tend to be cautious. Indeed, if it were your money, I'd probably be less cautious :). It's also an argument to be less trustful of these projections as well. Is that an argument in your favor? Is it an argument for the status quo? Status quo circa 19__?

What would make for a cautious course in this case? We don't know what will happen and it may be the case that Social Security will run short of money, so dismantle in now to spare us from having to dismantle it later?

That's a conspicuous nudge :). I don't pretend to have all the answers. Taking some pride in saying, I don't know. I would suggest small tweaks first. I like the idea of choice but that's simply on it's face and not backed by any scrutinized analysis. Pushing the retirement age a bit would be a first step in my opinion. We live longer now. We should work longer. My pet theory is an active mind fights off certain late-stage mental disorders and that seems to be a win-win here. Lowering late-stage healthcare cost and reducing the SS outlays.

Define "we" in "we live longer". Most of the evidence I've seen is that high-income people live longer (just one example: whereas everybody else lives about the same amount of time.

That's why I like the following proposal to raise the retirement age for those who live longer (i.e., higher-income groups), but not for others.

If we slice up social benefits like that, can we also slice up social premiums? Reduce high-income "Insurance Contributions" for Medicare's benefits since they are healthier? And reduce their FUTA contributions since they are less likely to ride benefits out for 73 weeks?

Or, metaphorically, does your ratchet only go lefty-loosey and not righty-tighty?

GKG, righty-regressy?

Regressive would be more benefits/lower costs to higher income people than for lower income people. GSG proposes lower costs corresponding to lower benefits. That's more like fairness than regressivity.

You can argue about which model is fairer, but most social insurance programs are designed to be at least somewhat progressive. They are structured such that lower-income people do not pay a (signifcantly) higher share of their total income than higher-earning folks for a certain benefit. If a benefits program and its funding design are not progressive, then they are by definition regressive.

Shorter Ryan: I take pride in not having all the answers, so I suggest we make old people work longer on the basis of my pet theory, not scrutinized analysis. Win-win!

No one is forcing you to work longer. Sorry to generalize, with the "we"'s folks. Indeed, I need to watch it. Thanks for the links Libert.

Social security ran short of money several years ago. We know very well what happened. We borrowed money to pay it, and people wanted to raise taxes to cover it.

why is it an argument for austerity?
also, if "progress" is only going to help a small number of people (unlike progress as it's been in the past) why would we let it happen?

Better safe than sorry on climate change.

It's 'safe' to spend hundreds of billions (excluding cost of regulations) on policies that have no discernible effect on temperature?

And despite the failure of the IPCC's modeled predictions, a wholly unexpected 16 year flatline in temperature and early reports that the next IPCC will significantly downgrade the sensitivity of climate to CO2.

More sorry than safe, I think.

Personally, I think Tyler is just having some fun here. There's no way anyone can get Tyler's interpretation of those comments unless they're as poorly informed as tenured faculty at the University of Chicago's Economics Department.

What I don't understand is this sudden outbreak of concern about increasing returns to owners of factors of production when the marginal output per unit rises. I'm pretty sure something like that is in Krugman's International Economics textbook somewhere. If anything this implies an argument in favor of higher taxes on dividends and capital gains.

How does a changing **share** of labor and capital in output result in an overestimate of the output gap by Keynesians? It seems that it would lead to an underestimate. E.g., if at full employment, labor is producing 60% instead of 65% of all output, given that we have a certain number of workers who produce a certain amount (say, all at minimum wage for simplicity), the GDP will be higher in the former case than the latter. So the output gap will be higher.

About 15 years ago I met a Chinese businessman from the Philippines at a party. I asked him what was the purpose of his visit. He said he was shopping for automated equipment for his consumer products manufacturing business back home. I said how does that work if you can get labour for 50 cents per hour. He said it doesn't and it will cost me extra money but I am so tired of coping with religious holidays and having to hire somebody's aunt just to keep the peace. I just want to minimize the number of employees I have to deal with.

The scope and rate of automation has been accelerating ever since. When are we going to notice that employment is eroding faster than the arctic ice? How many will still have a job in 2040?

Better safe than sorry? SS and AGW aren't even in the same ballpark.

SS is an intergenerational transfer program that most likely is insolvent because projected revenues won't cover expected outlays. You can argue about the models or actuarial tables, but the policy prescription is to shift the burden from one generation to the next, plain and simple. The public choice aspects aren't simple but the statement of the problem is simple and a solution is simple.

AGW on the other hand might not exist, and every measure taken to deal with a non-problem has efficiency as well as distribution problems. If we knew it existed, then it becomes a question of whether or not to deal with it and, if we do something, who will bear the costs. That involves an even tougher public choice problem of dealing with a public bad.

What's the "safe" way to fix SS? Do nothing until it explodes. The right thing to do is privatize it before every current working contributor gets screwed.


I am always amazed at the use of language to substitute for logic.

Take the term: "intergenerational transfer". Think about it for a moment. What does it really mean. But, first, let me put it in a different context, and then we'll come back to it.

Education is an intergenerational transfer. I pay for my kids education, I even pay for someone else's grandkids education (now since my kids have moved away). Their grandparents paid for mine.

I paid for a war; someone told me we would be worse off in the future if we did not fight it now...that future extends for quite a while; my kids got the transfer benefit.

Social security. My parents paid for theirs. As their kid, I paid for theirs. As my kid pays for mine. So long as each generation passes to the next, without someone like Bush doing an unfunded Part D, it all is a wash.

If you want to speak logically, say you are concerned about one generation increasing benefits for itself, and don't use the false coinage of intergenerational transfer, because it is all intergenerational transfer.

No, Bill, intergenerational transfer is not false coinage. The solvency of the transfer system depends crucially on longevity, income growth, and population growth cycles. If the system cannot tolerate swings in those factors much less growth in some relative to others, then it is a game of hot potato. Some generation in the future is going to get screwed.

Voting yourself benefit increases at the expense of future generations only makes the usual problems more severe. Redistribution of income through the benefit function makes it worse.

An education system likely would never go insolvent because classroom education is a public good, so population bubble effects can be mitigated. It has no longevity risk. Stagnant income growth isn't going to affect the feasibility of education. If things get really bad, home schooling is an option. There can be a market for education to lower its cost. SS doesn't have these features.

War is a public good too. Social security benefits are a private good. It is not, as people claim, an insurance system. People don't pay premiums associated with their risk of superannuation. If it did, the poor would have a 30% tax and high income earners would be exempt. It is an income redistribution system, plain and simple. The resemblance to a forced savings program is window dressing. If I could have, I would have opted out of SS more than 30 years ago and my retirement would be more secure because of it.

If I could identify the single biggest lapse in economic education and public policy debates, it would be this confusion of the difference between a public good and a publicly provided private good.


Please tell me, specifically, the facts supporting this statement: "Voting yourself benefit increases at the expense of future generations only makes the usual problems more severe. Redistribution of income through the benefit function makes it worse."

Other than Part D, please name the legislation and date of enactment where there was a vote for a benefit increase. I don't think SS has changed since 1983.


SS is much more terra firma than Krugman appreciates. Ironic given his arrogance about predictions about complex systems generally.

At current tax and benefit rates, the system is reasonably close to 'generationally fair'. This wasn't always the case. For the first 45 years of the program, FICA tax rates were lower and it was correct to speak of an 'intergenerational transfer', but for anyone who entered the workforce since the early 1980s, it's been close to fair and can be 'fixed' with undraconian steps:

Medicare, Medicaid, Obamacare- there's your tsunami.

+1 to Brian

Willitts has gotten very apocalyoptic lately...

'Apocalyptic'. Sorry.

A negative rate of return on my SS contributions, compared to the risk free rate and weighing the miniscule probability that I would have to rely on SS is NOT generationally fair. What are you talking about? Money was stolen from my wallet. I need police on the scene, not a claims adjuster.

What I'm talking about is whether or not Social Security is broken or fundamentally unstable- Ponzi-ish, if you will. My contention is that it basically is not, and the math and uncertainty behind Social Security is much less formidible than that for more open-ended promises like medical care.

You complain that you're not getting you're "money's worth". Well, there are subsidies in Social Security, from (for example) high-paid to low-paid (because of the way the benefit formula works), or in-favor of 'traditional' families (non-working spouse benefit.) There are also insurance "winners" (those who die with dependants, become disabled, or live a long time) and losers (those who croak just before they get their first check.)

All of this is 'intra-generational' though. The fact that YOU (and me) pay in more than we get out is evidence that the system is not Ponzi-ish.

And don't underrate the value of the benefit. Go try and buy an inflation-protected annuity out there in the real world these days. Based on today's yield curve and the prospects for monetary policy, I don't believe there are any investments out there that can be confident of beating inflation over the next decade. I think a lot of people spend a lot of their careers complaining about Social Secuirty, then they get to retirement and they're like "Really? Not bad."

I'm Libertarianish, but I believe Social Security is the one thing the government has done that has worked. Behavioral economics before it was fashionable.

God knows the public finances in this country are a mess, but Social Security is a red herring.

" The fact that YOU (and me) pay in more than we get out is evidence that the system is not Ponzi-ish."



You might want to view the lectures on social insurance by Prof. Shiller at Yale on the subject of both social security and social insurance in general. Go to and look up his free course on finance. He'll go through the math Bottom line: you are purchasing an annuity, paying it over a lifetime beginning at around 20 and ending at 65 to cover a later time in your life. Not only that, you are getting disability insurance as well for your 45 year worklife. Because it is a compulsory pool that is not adverse selected you are getting something a market wouldn/t be able to offer--not because of the cost, but because of adverse selection, information assymetry, selling costs, etc.

Actually, the rise of the robots is an argument for higher taxes on capital.

Robots are going to eliminate work, although not right away. The nascent issue is "what does human culture look like when 'work' is voluntary, rather than the root source of human dignity."

Will we try socialism again for awhile, or move to leaving stuff in piles, and allowing people to take as much as they want, because the robots can easily make more? [I read today that the stronger-than-steel silk from spiders is now real.]

Comments for this post are closed