Permanent income hypothesis
In this brief post, Krugman clearly restates his clear (and plausible) view that rule of thumb ad hoc paleo Keyensians had a lot to learn from Milton Friedman.
He mentions the Phillips curve. I think I detect a hint of Forder in this sentence “We can argue how many economists really believed in a stable tradeoff between inflation and unemployment, but that’s certainly what got taught to many students.” But, this time, I want to write about the Permanent income hypothesis.
Krugman wrote:
Conventional Keynesian consumption functions suggested that the savings rate would rise as incomes rose — and this wasn’t just the Keynesian interpreters, Keynes himself made the same claim. This, in turn, led to predictions of rising savings rates after World War II, and hence a persistent shortage of demand — hence the secular stagnation theory briefly prominent. (There was even an early Heinlein novel built in part around the secular stagnation theory. As I recall, it was pretty bad.)
In fact, however, savings rates don’t seem to follow the naive consumption function at all; they rise in booms, and are higher for the wealthy, but exhibit no secular trend. And Milton Friedman appeared to explain this paradox by arguing that people are more less rational: they base consumption on “permanent income”, a reasonable estimate of long-run income, and save temporary fluctuations in income.
I have three thoughts.
First the patterns (looks like IS-LM Hicks in the cross section and short term and looks like consumer theory Hicks in the long term) can be explained by habit formation, even if consumers are myopic. Macroeconomists who try to fit aggregate data have been forced to assume habit formation. As far as I know (not as far as I should know) once on accepts habit formation, one can’t reject Keynes’ hypothesis that shifting expectations about future *aggregate* income have negligible effects on current consumption.
There is a clear pattern based on survey data, that if people expect rapid GDP growth then there is likely to be rapid growth of consumption over the same period. Such predictable changes in consumption are inconstent with the simplest crudest models used to illustrate the PIH. They can be reconciled with utility maximization (anything can be) but they are not at all the pattern implied by the PIH. The PIH implies that if high GDP growth is expected from time t on, then consumption at time t is higher than one would guess given income at t, wealth at t and lagged consuption. The PIH implies a relationship between expected future growth and the level of consumption. Keynes’ clearly stated hypothesis was that expected future aggregate growth was negligible
The principal objective factors which influence the propensity to consume appear to be the following:
[skip]
(6) Changes in expectations of the relation between the present and the future level of income. — We must catalogue this factor for the sake of formal completeness. But, whilst it may affect considerably a particular individual’s propensity to consume, it is likely to average out for the community as a whole. Moreover, it is a matter about which there is, as a rule, too much uncertainty for it to exert much influence.
As far as I know, the last sentence is a null hypothesis which hasn’t been rejected by the data. This is a bleg/challenge — links in comments to rejection of Keynes’s hypothesis will be much appreciated.
Second IS-LM if taken literally implies that there is no trend in consumption or GDP. It was introduced as a way of thinking about deviations from trends. I see no more reason to take the assumption of constant autonomous consumption literally than to take the implication of constant GDP literally.
Finally cointigration (sorry to technospeak it roughly means the same trend) of consumption and GDP can be explained by a myopic consumption rule which depends on lagged consumption, current income and current financial wealth. To avoid it, one needs to assume that consumption doesn’t rise with financial wealth. Now it is a bit tricky to reconcile paleo-Keynesianism with massive increases of the ratio household debt to GDP. But it is at least equally tricky to reconcile that increase with intertemporal optimization. 21st century US data sure seem to fit a rule of thumb involving consumption out of wealth (especially including home equity) a lot better than rational intertemporal optimization.
As far as I know, the null that the PIH glass is totally empty has not yet been rejected.
i am not sure i understand any of that, or if it matters. but here is an anecdata for you.
we didn’t have a lot of money when i was a kid, though we were never in “want.” if you had asked me at any time when i was a wage earner i would have told you i didn’t spend a lot of money because you never know when you might need it. and in fact i probably did save more than most people. but in fact when i did have a lot of money coming in and no immediate expectation of not having enough, i spent “more than usual” for things that were not necessary. privation, even self induced, is a strain, and one lets go from time to time. but note there was no “expectation” of continued high earnings, and certainly no “lifetime expectations.”
i suspect the theory doesn’t work too well because the people making it and using it have no idea how ordinary people think.
i will admit that the spending beyond our means (of people, not governments) that i see makes me mutter “no good will come of this.”
oh, yes: the savings rate rises as incomes rise. well, yes. when you have more money than you need for rent and groceries, you do tend to put some in the bank. at least I do.
Ok, too much uncertainty about the long run needs to be eliminated or PIH needs to be gone, they are exclusive? That is my interpretation. See Roger Farmer, i think he made the assumption that consumer acts like todays GDP growth will be repeated tomorrow. That is PIH? (i have to decode stuff i miss). So, Farmer lets labor equilibrium a free variable. More emphasis on the recent past is PIH. He is close to proving the one over the other.
Keynes wants long and short separable. That is very difficult to observe.
I can give an example where keynes is right. A economic model of five square miles of apartments in LA. A two good model, apartment rentals and apartment maintainence. Entrys equal exits, all transactions independent and fixed in time, andd symmetrical movements in and out. Equilibrium bell curve, efficiency gained be making bell sharply peaked. Bias center of curve at zero without loss of generality, assume continuous.
X axis of curve is 1/(rental period) ok, no short term tenant meets any long term tenant until contract complete. I think i have it all.
External shock hits, flooding and damage on street. More tenants leave than enter. The distribution of ‘ghost rentals’ rises, bell slides in negative direction. But apartment maintainers save the day getting repairs done while road is repaired.
Proof that Keynes was right at least once.
Got it. Milt and Keynes had the same belief, the idea that the consumer independently looks at short vs long. I missed that, sorry.
So then, Farmer argues against, Farmer says the consumer acts as if the recent gdp repeats.
No. The consumer has a portfolio, an inventory, which is finite and low dimensional. The consumer maintains that inventory out of available market transactions. Typical inventory is food, clothing, car, housing, leisure, kids and spouse. Small, low dimensional, finite and all of them co-related, none orthogonal. Cost of portfolio adjustment is high, user hesitates to incur cost. This is the permanent purchase hypothesis, the correct one and obey Farmers discovery.
“Now it is a bit tricky to reconcile paleo-Keynesianism with massive increases of the ratio household debt to GDP.”
It’s possible to do so if you take into account credit constraints. The aggregate consumption function reflects a mix of savers and borrowers. In general there are always more people wanting to borrow than are considered creditworthy. The consumption function therefore always shows a higher aggregate savings ratio than would apply if everyone could borrow what they want. If lenders relax their lending criteria, the overall savings ratio will fall.
This has certainly been a factor in rising debt ratios, although I wouldn’t say it explained it all.
@Nick Edmonds I agree that is the sort of “a bit tricky” explanation which is probably needed. The idea that aggregate saving rates have a lot to do with credit constraints is strongly supported by international data — where credit is easy (key case the USA) saving is low.
I think there are also a short term fluctuations due to asset price bubbles. The PIH assumes that people are roughly rational so such bubbles are impossible. Keynes other paleo Keynesians and all sensible people thought and think that they are a key part of the business cycle.
like i said, i probably don’t understand this. what i take away is that some good observations are worth talking about, but “economists” go wrong when they try to promote one of those observations to a universal law… and predict things like “bubbles are impossible.”
I’m not sure that this is directly & strictly on topic… But I’m with Coberly on the issue that a lot of the ideas some have seem to be ignorant of the situations of a lot of ordinary people. Or at least just seem to ignore the reality of a lot of ordinary people.
I’ve seen often this idea that people will spend more of their current income, based on an expectation that their income is going to rise in the future (presumably fairly near future).
I think it’s true that there’s a lot of pent up necessity demand among ordinary folks right now. In that people hold off on needed things and do without because their income is too low or too insecure. And when they get the chance, they of course spend more.
However, the idea that people will spend more now, expecting an increase in the future, assumes that people already have excess income, savings, or access to comfortable credit, that they’re willing to use.
Do they have these and are they willing to count their chickens before they hatch nowadays?