Permanent Income Hypothesis
By Matthew Yglesias
Megan McArdle writes about a potential problem with stimuli:
The real question, I think, is how close the permanent income hypothesis is to being true. The basic idea is that people are forward looking, and they try to smooth their consumption over time. So if you give them a “temporary tax cut”, they save most of it, knowing that eventually they will have to give the money back.
But of course, this should also be true of “temporary government spending”–if people think the money won’t be there next year, they’ll salt as much of the money away as possible. This is a topic very underexplored in the various estimates of the stimulus multiplier, even though consumers are massively overleveraged and will presumably save as much of their new income as they can.
I think common sense tells us that the permanent income hypothesis is very far from being true. You could imagine an alien species whose psychology functioned this way. But that species wouldn’t have heavy smokers dying of cancer, problems with overreating and sedentary lifestyles. Voters belonging to that species would condemn governors who take advantage of boom times to cut taxes and hike spending—there would be massive popular pressure to sock it all away in a rainy day fund.
It’s clearly true that in the present situation people are saving more and spending less. But the reason can’t be that they’re perfectly forward-looking—the evidence for high rates of time preference and myopia is too overwhelming.
60 Responses to “Permanent Income Hypothesis”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
January 31st, 2009 at 1:01 pm
Megan seems to be confusing the Permanent Income Hypothesis with Ricardian Equivalence. Either way, I wouldn’t say they’re being overlooked in the multiplier discussion since the measurement of multipliers is an empirical matter.
January 31st, 2009 at 1:06 pm
The real question, I think, is how close the Megan’s Skull Filled With Cement hypothesis is to being true.
January 31st, 2009 at 1:06 pm
Explain again of all the bloggers in the world you cite Megan so often?
Surely there must be bloggers with more interesting ideas you could critique?
January 31st, 2009 at 1:07 pm
There are lots of papers on this topic and the consensus is that deviations from the PIH are substantial for non wealthy households. One nice recent paper is David S Johnson,Jonathan A Parker, Nicholas S Souleles Household Expenditure and the Income Tax Rebates of 2001 (AER 2005)
One caution thought: the PIH is a good baseline for understanding the world, and people don’t always deviate from the PIH in ways that are easy to understand, stable or which make policy easy.
January 31st, 2009 at 1:09 pm
Wow there is so much wrong in those two paragraphs you’d think McArdle was Bill Kristol. As stated she mixes up PIH with Ricardian Equivalence. She seems to think this hasn’t been looked at when it has. But hey when she studied with people as out to lunch on teh stimulus as the Chicago guys it isn’t surprising.
January 31st, 2009 at 1:14 pm
I don’t believe that saving during a recession is as terrible as Keynesians think it is, or that spending is as great.
If wages were allowed to deflate during recessions, I think that much of the unemployment would be solved and many of the problems relating to reduced prices would also be solved.
The theory is that without “sufficient demand,” prices fall and so businesses must cut jobs, etc. But if wages were to fall to levels commensurate with the new prices, unemployment would be much less, and everything would get attuned to the new price system.
The one problem, of course, would be people who owed money, but this would be better solved by renegotiating debts and forgiving portions of debts that are too high in the new deflated currency. Ultimately, as recessions are caused largely by unsustainable expansions of credit, some increase in debt defaults and/or debt forgiveness is inevitable and necessary, to get rid of unsustainable loans and bad business models.
January 31st, 2009 at 1:15 pm
It’s interesting: many otherwise thoughtful liberal bloggers treat McArdle like a decent thinker and writer who just happens to have a few eccentricities/blind spots. It just isn’t so: she’s wrong about almost everything.
January 31st, 2009 at 1:18 pm
As any econ grad student knows, the Permanent Income Hypothesis doe not apply, even under pure forward-looking rationality, when people are credit-constrained – when they do not have access to borrowing. Americans, needless to say, are more credit-constrained now than anytime in recent memory.
Megan McArdle would know this if she had ever, say, studied economics.
January 31st, 2009 at 1:23 pm
Matt, once again you have demeaned yourself by linking to that charlatan.
January 31st, 2009 at 1:26 pm
Longtime reader and fan of your blog but I implore you to stop linking and commenting Megan McArdle. She has zero credibility and no insightful thoughts. There are blogs dedicated to highlighting her terrible writings.
You are doing yourself and your readers a disservice by providing her a platform. This is only hurting your credibility.
January 31st, 2009 at 1:28 pm
You guys realize that at this point he’s linking to McMegan just to get a rise out of you, right?
January 31st, 2009 at 1:32 pm
I hate to inform you all of this, but, homo-economicus does NOT exist.
January 31st, 2009 at 1:34 pm
I’m beginning to suspect that about half of Mr. Yglesias’ posts are intentionally designed to piss off his commenters. And I love him for it.
January 31st, 2009 at 1:36 pm
You guys realize that at this point he’s linking to McMegan just to get a rise out of you, right?
And it’s that kind of log-rolling and borderline trolling that will propel MeMeMe to the op-ed slot she so clearly thinks she deserves.
January 31st, 2009 at 1:45 pm
I was gonna post about how megan didn’t know what she was talking about (again) and how Matt should really stop addressing her vapid, silly pseudo-smart posts, but then several people already did it.
Matt are you two friends or something and you feel bad for this silly dim-wit, so you figure some links will help her out?
January 31st, 2009 at 1:45 pm
Megan McArdle is patently stupid. Every time her name gets mentioned, it’s like giving publicity to the flat earth society; surely we could be talking about serious things – and people – instead of nonsense, but i guess that is just the Washington way. Continue…
January 31st, 2009 at 1:46 pm
I couldn’t get past “Megan McArdle writes.”
January 31st, 2009 at 1:47 pm
Is she arguing against tax cuts in the stimulus bill?
January 31st, 2009 at 1:51 pm
Teh Megan writes (emphasis added):
What does this word “save” mean? I don’t know anyone who is saving anything. I do know people who are paying down their home equity lines and credit card debt so that they’ll have a credit buffer they can use again if necessary, and trying to pay down their mortgages so they won’t lose their homes in the coming depression. But people are “saving” their tax breaks? Perhaps in Meganland. Not in Kansachusetts.
I agree with those who say you should stop trying to cure — I mean, teach — Megan, Matt. She is not of this world.
January 31st, 2009 at 2:01 pm
I’m beginning to suspect that about half of Mr. Yglesias’ posts are intentionally designed to piss off his commenters. And I love him for it.
You must love a lot of pundits/commentators these days, then, since—to put it mildly—it’s not real hard lately to find bloggers/columnists/TV talking-heads putting out aggressively half-baked contrarianism in hopes of pissing rational people off.
January 31st, 2009 at 2:02 pm
Matt, you should’ve looked up PIH before you blasted into McMegan about it. Then at least you would have realized just how comically wrong she is. Incidently, this mistake she made, it’s a common mistake students make when they’re exposed to PIH for the first time.
In short, PIH is closely related to Ricardian Equivalence, and RE applies only to the taxation side of the problem. In the neoclassical model, it is entirely possible to grow the economy through increased spending (various forms of public goods). PIH says in this case only that it doesn’t matter how you fund these goods, all that matters is the initial wealth of the consumer and the time path that these goods will be provided.
January 31st, 2009 at 2:12 pm
The permanent income hypothesis seems have some explanatory power within my life experience – at least for those of lower-middle-class expectations and above (a shrinking group to be sure).
However, people do have quite a bit of power of day-to-day observation, and right now most people I know (of every class except Cheney-wealthy) are anticipating that every dollar they spend over and above survival requirements brings them one day closer to living in a cardboard box under an Interstate overpass. Such immediacy is clearly going to overwhelm any subtle long-term psychological effects.
Cranky
January 31st, 2009 at 3:18 pm
What does this word “save” mean? I don’t know anyone who is saving anything. I do know people who are paying down their home equity lines and credit card debt so that they’ll have a credit buffer they can use again if necessary, and trying to pay down their mortgages so they won’t lose their homes in the coming depression. But people are “saving” their tax breaks? Perhaps in Meganland. Not in Kansachusetts.
I would not defend McMegan, but in what sense is paying down debt different from saving? In both cases net worth increases. There is no reason to treat them any differently and the BEA does not treat them differently in their statistics about consumer finances.
January 31st, 2009 at 3:18 pm
Huh? These arguments don’t make sense to me at all.
1) I think it’s very well-established at this point that the psychology of addictive drugs and the psychology of eating are completely different from the psychology of buying consumption goods vs. saving. The vast majority of smokers start in childhood and then find it very difficult to quit because nicotine is addictive. Likewise, people overeat because humans are evolutionarily hardwired to eat a lot.
2) I agree that there’s little tendency to support rainy day funds, but what usually happens in bad times is that similar levels of government spending are supported by running bigger deficits, the interest on which is paid back later. (Many states are required to have “balanced budgets,” but they effectively run imbalanced budgets by having voters approve bond measures).
January 31st, 2009 at 3:21 pm
I would not defend McMegan, but in what sense is paying down debt different from saving? In both cases net worth increases. There is no reason to treat them any differently and the BEA does not treat them differently in their statistics about consumer finances.
For PIH purposes, paying down debt is exactly the same as saving.
January 31st, 2009 at 3:28 pm
no comment, I was replying to Kansachusetts. I agree with you – the question was rhetorical.
January 31st, 2009 at 4:32 pm
Re: I think it’s very well-established at this point that the psychology of addictive drugs and the psychology of eating are completely different from the psychology of buying consumption goods vs. saving.
It’s also very established that human beings have limited time horizons: anything beyond a couple of years is unreal to most people. One need not invoke addictive drugs to see this. Why do so many people fail to save for retirement? Have children they cannot support? Engage in unprotected sex? Not maintain their vehicles properly? For that matter, why do businesses focus laser-like on the next quarterly statement and scant long-term profitability? Heck, why did the mortgage industry let itself go to hell in a handcart? If you put reward or punishment too far off in the future it drops off the human radar altogether.
January 31st, 2009 at 4:40 pm
Why do so many people fail to save for retirement?
About 80% of people save enough for retirement.
http://www.slate.com/id/2193350/
(The paper is from 2005. The hit on the stock market in 2008 probably hurt lots of people about to retire.)
January 31st, 2009 at 5:16 pm
I would recommend “The Selfish Gene” by Dawkins for anyone wondering why people act and think the way they do.
January 31st, 2009 at 5:23 pm
Noah gets bonus points for bringing up the point about credit constraints.
January 31st, 2009 at 6:12 pm
Hey, puzzled and no comment agree: For PIH purposes, paying down debt is exactly the same as saving.
OK, I’m saving! I’m saving! That’s a big relief, because I thought I was just getting myself into a position where I could borrow some more money at whatever the prevailing interest rate turns out to be once I get laid off.
I’m sure you guys are right about the PIH definition. For some reason, though, an economy where lots of people are paying down debt does not strike me as quite as healthy as an economy in which people are actually saving or investing real money. If that is not at all germane to the argument, well never mind.
January 31st, 2009 at 6:43 pm
Matt,
Any thoughts on Christopher Caldwell’s column in today’s Financial Times (“Is the stimulus Obama’s Iraq?”)? That may be an analogy too far, but Caldwell makes some worthwhile points. Excerpt:
January 31st, 2009 at 7:58 pm
Gosh, it would be wonderful to live in a world where, when you got a tax rebate, you could logically decide to put it in a savings account paying maybe 5% interest, instead of buying food and clothing, or paying off a credit card bill that’s charging you 27% interest. I guess that would be the world of cute young women who inherited money and got sent to a name-brand school, so they’re “in with the in-crowd”, as the old song said.
In the real world, in case the Ivy League hasn’t noticed, consumer sales are alternating between flatlining and heading for the second basement.
Also in the real world, there are plenty of people who don’t stuff themselves with food until they’re twice normal size, in spite of the fact they could afford to do so. They’re called Europeans.
January 31st, 2009 at 8:48 pm
Part of the reason why wages are “sticky” is because there is an effort to make them so.
During the beginning of the Great Depression, Hoover campaigned hard toward businesses that they not cut wages at all, with the result being unemployment, and a worsening of the Depression. In fact, part of the reason for the credit expansion of the 20s which is in the Austrian view the cause of the recession that became the Great Depression was because Britain wanted to keep union wages high at the same time as it was pursuing deflationary policies, so in an effort to be friendly the U.S. pursued inflation to get Britain the extra money.
The problem is that this can become a cycle, not a one-time move to a new equilibrium. Prices fall, production goes down, then wages are cut. Consumer demand goes down, so prices are cut even more, production goes down even more, and wages are cut even more. Consumer demand goes down even more, so prices are cut yet again. And so on.
But eventually those who have a lot of savings are able to buy lots of stuff at greatly reduced prices, causing the cycle to end.
Second, just because an excessive expansion of credit may have helped cause a recession (as it arguably did in this case), that doesn’t mean that a excessive contraction of credit is a good cure for a recession.
The question though is how much contraction is excessive, and if the government does not let an appropriate amount of contraciton take place, do they not risk exacerbating the credit bubble?
January 31st, 2009 at 9:03 pm
Although it does create some moral hazard problems, in my opinion arguably the best stimulus (if we are to have a stimulus package) would be to have tax credits or tax deductions or subsidies, etc. designed to be skewed toward the people most likely to be in debt over their heads, and skewed toward being used to pay off those debts rather than toward increasing consumption.
People keep arguing that the best stimulus would be weighted to the poor because they are most likely to spend it. I think a fair argument could be made that the best stimulus would be weighted to those in the worst debt situations (which would likely skew heavily toward the poor or at least the lower-middle class), because they would be the most likely to retire the most dangerous debt.
This would help to solve the problem of banks not lending, if they were more convinced that they would get paid back. And it would avoid the issue that others have mentioned, that paying off debts won’t help the credit situation because it would be the highest-rated debt that gets paid off, because the skew would deliberately be toward the lowest-rated, most “toxic” debt.
And the concern that retiring debt would be deflationary and that a reduced debt load would somehow hurt the economy would also be avoided, or at least greatly ameliorated, because many of the debts that get partially retired were going to end in default anyway, so thedeflation would have happened in any case. And don’t forget the default multiplier effect – by helping to retire a portion of the value of the house, you prevent default on the entire value of the house (minus whatever the bank can salvage in the foreclosure), so the overall effect could be non-deflationary.
January 31st, 2009 at 9:05 pm
So DTM, what is your opinion on a stimulus measure designed to help people who on the margins of default to retire debt (i.e. not people who are way in over their heads who can’t be saved or those who can continue to finance their current debt load).
Not, I note, to help them necessarily to completely retire their debts, but to retire enough to keep their heads reasonably above water.
February 1st, 2009 at 12:23 am
So DTM, you are partly admitting a point I made a while back.
The idea that we need to “stimulate” the economy by giving money to people who will spend it right away (i.e. consume) rather than to people who will save it or use it to retire debt, even at a time when so many are drowning in debt, is because you don’t see all of the excess debt as a bad thing because you don’t intend for a lot of it to be paid back.
Maybe not on the federal government level, but definitiely you feel this on the individual level.
February 1st, 2009 at 8:34 am
McArdle is totally wrong about the implications of the PIH as noted by DeLong and Krugman.
The PIH implies that tax cuts do not stimulate the economy. It implies that temporary increases in spending stimulate the economy. One can argue that a permanent increase in public spending will not cause increased demand (it depends on whether the spending is productive and on aggregate supply). As Krugman notes she got it exactly backwards.
It is, however, possible to reconcile smoking with the PIH so long as you assume it is real fun. No actual smokers find this remotely plausible, but Gary Becker and Kevin Murphy felt the need to point out that it is possible to model a rational agent with dynamically consistent preferences who consumes an addictive substance. Oddly, they don’t seem to have noticed that no one ever claims that the fact that people drink coffee is proof of irrationality, nor the fact that many people who say that smoking is irrational used to smoke (typically I think until about one week to one month before they wrote the article about smoking because they couldn’t think about anything else).
February 1st, 2009 at 8:48 am
In general, you seem to struggle with the notion that economic policy has to deal primarily with the present and the future, not the past.
No, my point is that you seem to feel that the way to get us out of a recession caused by a credit crisis, which is ultimately caused by overconsumption, is to take the hair of the dog that bit us and get people to consume even more.
We got in this problem because we spent beyond our means and the Keynesian solution is to try to avoid the correction by having the government “stimulate” the economy by spending beyond its means and encouraged us to spend some more.
You have this exactly backward. From a deflationary perspective, a default on a debt is fine (at least in terms of first order effects): whatever money was going to pay back the debt is now free for consumption instead.
I was under the impression that in a fractional reserve system, money was created through debt, and that a default or write-down on debt essentially destroyed the money generated by the debt.
I guess our real difference is that I do not believe that recessions are caused by a crisis in demand per se. Rather, I believe that they are caused by a mis-coordination of investment and good production. In the 1990s, the tech bubble grew because easy money allowed people to bid up the prices of stocks and to make money speculating without receiving ny dividends. The bubble popped when it became clear that most of the new companies did not have a sustainable business model and were simply being sustained through one loan after another as they tried to buy up enough of the diminishing resource base to stay in business. Essentially, too much money was chasing too few goods of production.
The response was more easy credit and money expansion, which created the housing bubble.
My point is that if we respond with massive government consumptive spending or with more easy credit, we will simply set off another bubble and will have the same problem come back in short order and the new bubble pops.
What we need is to increase right now is not consumption (”demand”) but to invest money in increasing our productivity so that we can produce enough to finance our standard of living.
While I do not subscribe to the theory that government infrastructure projects are the best way to do that, I think that such projects are far superior to attempts to get consumers to spend more. Anything that involves increasing our investement in things that will lead to more productivty (better internet connections, better roads, upgrading air traffic control hardware and software) will make us more productive and help our lifestyle to be more sustainable. Things that simply encourage us to consume will not. (It should be noted then that make-work government projects, such as making scenic bike trails or beautifying paths are not really that helpful from a stimulus perspective. We need things that can be justified on the basis of actual practical usefulness).
February 2nd, 2009 at 9:58 am
A loan is an asset for a bank; it shows up on the asset side of the bank’s balance sheet. A deposit (by which “money is created”) is a liability for a bank; it shows up on the liability side of the bank’s balance sheet. Thus, deposits are the form of bank debt that “creates money” not bank lending (which is a debt to the borrower but an asset to the bank) A bank’s assets (which may include some loan loss reserves) by accounting definition equals its liabilities plus and its capital.
A loan write off comes off the asset side of the bank, with there not necessarily being any reduction in the liability account (the “money creating” account) — if there were sufficient deposits (from the borrower) on the ledger to compensate for the write off amount, there would be loan repayment, not “loan write off.” Thus, when there is a loan write off, either loss reserves are drawn down or bank capital is reduced.
So all this “writing down assets destroys money” stuff seems kind of misdirected or worse. In fact, deposit insurance exists precisely because loss of asset value does not necessarily destroy the depositor claims on the bank.
I think Noah above has it right about the lack of credit constraints being fundamental to the PIH story.
March 1st, 2009 at 4:58 am
viagra
I bookmarked this site. Thank you for good job!
March 11th, 2009 at 4:11 am
Incredible site!
March 12th, 2009 at 10:39 pm
I bookmarked this site. Thank you for good job!
March 14th, 2009 at 4:55 am
I want to say – thank you for this!
xanax
March 22nd, 2009 at 5:54 am
tramadol
Excellent site. It was pleasant to me.
March 22nd, 2009 at 10:18 am
buy viagra online
It is the coolest site,keep so!
March 23rd, 2009 at 4:18 am
viagra
I bookmarked this site. Thank you for good job!
April 2nd, 2009 at 4:51 am
It is the coolest site,keep so!
buy cheap viagra
April 3rd, 2009 at 3:58 am
Excellent site. It was pleasant to me.
cheap brand pfizer viagra
April 9th, 2009 at 4:39 am
thanks !! very helpful post! viagra