Empirical reality has told us that–at least when inflation is very low, as it is at present–the short-run is not less than five years but (shudder) can be as long as fifteen.
The full post, which offers more, is here.
That is exactly the kind of direct response I have been looking for, though I might get greedy and ask what makes inflation “very low.” Core inflation has now reached two percent and I can’t quite regard the non-core, which is higher at 3.8 percent, as totally irrelevant. (Why is it I hear Scott Sumner in my ear, and can you guess which four-letter abbreviation he is screaming out?)
In my view, supply-side factors are the main reason why the employment-to-population ratio has been so dismal since 2000, demand-side factors are the main reason why so many bad things have happened since 2007-2009, and supply-side factors and mismatched expectations are the fundamental reason why demand-side factors went south in 2007-2009.
It also seems to me that the long run comes more quickly when TFP is relatively high, which again brings us back, at least partially, to the supply side. This view is supported by theory. When the economy has a lot of broad-based technological innovation, at least somewhat evenly distributed, job creation is easier, income effects are more likely to positively cumulate, and monetary and fiscal policy are more likely to gain traction.
Part of me is willing to accept a linguistic bargain, something like the following: “I will admit that the short run can last for fifteen years, if we agree that TFP helps determine that horizon.” Another part of me then realizes that if the long run helps determine the relevance of the “short run,” the long run is always mattering. At which point I go back to believing that the traditional “old Keynesian” distinction between the long run and the short run is sometimes more confusing than illuminating.
Sorry but I prefer my Argentine rule of how short is a short-run. At the beginning inflation was 5% every five years, then it was 5% every two years, then 5% every year, then 5% every six months, and then accelerated to 5% every month, to 5% every week, to every day, and then it disappeared because we stop using the peso.
If you want to be serious, go back to Marshall and then combine and expand Walter Oi’s and Arnold Kling’s ideas to argue for two types of labor: one as a fixed factor and the other as a quasi-variable factor. Key question: How can wages and benefits to the two types of labor include fixed and variable payments?
TIPS!
Oh yes, mine are doing very well, though they were 30 year ones bought back in the late 1990s (silly me, I didn’t buy stocks back then) – paying around 4%, plus the principal adjustment having added another 30% or so to the principal (which just increases the amount of money earning 4%). What a fantastically stupid way for a government to finance itself, though those ‘lending’ the money to the government are more than happy with the deal – after all, a (miniscule) portion of taxes paid by Americans are actually my revenue. This mechanism is one of the fundamental reasons that a political economy ends up so skewed – those at the botttom blame government for being taxed, while those receiving a significant porition of that tax revenue (look at America’s debt service to understand the amount of ‘transfer payment’ there is from taxpayer to bondholders) have zero incentive to reduce their revenue stream, as ensured through taxation. One side is under no illusion how this game is played – another reason why the rich are different, when viewed as a class.
However, purely theoretically, if true deflation were to occur, the amount of inflation adjusted principal would actually decrease to reflect that amount of deflation.
When the economy has a lot of broad-based technological innovation, at least somewhat evenly distributed, job creation is easier, income effects are more likely to positively cumulate, and monetary and fiscal policy are more likely to gain traction
Reminds me of the good old days of Greenspan’s virtuous cycle.
I dont see any “definition” of the long run anywhere in there.
In Micro, the definition of the long run is a period of time sufficient for the firm to vary the quantity of all inputs.
In Macro, the long run is defined as the period of time sufficient for prices and wages to adjust to return e economy to full employment. But FE is a moving target, affected both by individual and policy decisions. Remedial policies intend to shorten the long run, but they often have the unintended consequence of lengthening it.
This downturn was characterized by enormous investment in and consumption of durable goods which, by definition, suppresses demand for inputs and output for many years. We also incurred enormous debt to finance this. All the current policy prescriptions are to create more durable goods and incur more debt. Simply brilliant!
Demand has been satiated for years, and there’s still a glut of supply. Government is doing everything it can to pass the pain into the future and take it off the shoulders of those who rightly deserve it.
BDL wastes no time dragging the red herring of income inequality as if it has any economic significance and as if we have some moral imperative to make things right through redistribution. Then he wraps up his argument with a dehumanizing description of his opponents.
How typical.
Income inequality has no economic significance? I know economists tend to ignore it and conservatives pretend our Gini coefficient was handed to us by God, but this statement is ridiculous.
Then what is its economic significance?
High inequality lowers distributive efficiency and aggregate demand. I’d bet it encourages asset bubbles and destabilizes the financial system. Do you believe an economy with the bulk of it’s resources controlled by a small number of people will behave like one with a more even distribution.
Agree with Lars, without disagreeing with Mike as to consumer “investment” in durables and the concommitant long term debt workoff as suppressing private consumption, but I do not agree that government, if it is purchasing now what it would have to purchase later, is necessarily putting anything on the back of future generations. Rather, young people today need jobs so that they can become even more productive in the future, and denying them that first job–be it in government or be it in a private contractor who bid for a government contract–is to waste them for no good reason.
That’s the damage to the future generation.
A spike in inequality would lower aggregate demand in the short-run, but it wouldn’t matter in the long run. And a spike large enough to generate a business cycle is highly highly implausible. And in recessions like the recent one where the stock market gets hammered the Gini tends to fall or at least stay flat, as the wealthy have a greater portion of their income tied to bonuses or stock options and a greater portion of their assets in the stock market.
And what do you mean by distributive efficiency?
What, pray tell, is “distributive efficiency?” The only definition of that I’m familiar with is whether markets clear. Unless your definition ends with “to each according to his need,” then your meaning is quite clear.
Even if we assumed equal endowments and income, capital would pool according to its moat efficient use. There would be winners and losers, resulting in different economic outcomes not necessarily the result of random chance.
The managers of large pools of capital will still possess the “commanding heights.” Industries with economies of scale and those willing to assume risk will end up larger and richer than those which do not. The difference between your view and mine is that i want the chips to fall where they may, and you want to choose the winners and losers.
There is plenty of research that showed economic development resulted precisely from unequal distribution of income. Only those with surplus income could afford or were willing to accept risk for capital intensive ventures. In modern financial markets, the smallest income earner plays a role and has a stake in economic growth. The world’s largest bank pays interest on loans made to it in the savings accounts of paper boys.
No, it has no economic significance whatsover. It possesses only social significance to those who think a transfer will alleviate both suffering and productivity. I don’t accept that view.
Someone creating wealth through innovation and reaping the rewards either as entrepreneur, manager, or worker does not deprive me of anything whether I am employed or not, nor whether i transact with that firm or not.
Rising income inequality is a statistic without significance, at least as far as our economy is concerned. Rising income inequality in China is remarkable. Their Ginos is higher than ours. Ginos also doesn’t capture mobility along that line.
When the economy has a lot of broad-based technological innovation, at least somewhat evenly distributed, job creation is easier …
Net job creation? Why is this necessarily true? Does a low per capita growth rate necessarily imply high unemployment, or am I misunderstanding Cowen?
Hmm, I guess I should’ve put in a paragraph tag.
Isn’t it scary that we now think we know what ‘short run’ means because DeLong says so?
No wonder Krugman and all his minions think they are right. He who defines the rules of the game usually wins.
Don’t worry. They are winning only backyard games. Apparently, in his new book Ron Suskind argues that Larry Summers was a one-player football team –he liked to throw the pass and receive it. I assume that post-Larry there has been no team –and no game, just circus.
I think most economists operating today should be put through a boot camp run by a bunch of Russian physicists.
When I say long time behavior is not relevant for some short time scale, I say that and write T >> t representing an approximation to a function f(t) that ignores terms of o((t/T)^n) for some n. A Taylor expansion. I could also write that in reverse given a short fluctuation timescale t0 << t, and then I could talk about t0 << t <0 out in the fit to the long run trend.
The other possibility is that g(t) depends on some other timescale T’ >> t0 that defines a separate separation between a new long run and a new short run in which case you and Brad Delong are talking past each other. He is talking about long run of macroeconomic variables T where GDP follows a trendline for times ~ T and you are talking about long run of some other mesoscale effect in which case you should probably refer to it as the medium run and Brad Delong et al are just lumping your effects into the macroeconomic short run.
N.b. t0 here is longer than the hour-to-hour market fluctuations that are probably controlled by the EMH.
There was a typo/mangling:
… I could talk about t0 << t << T in the fit to the long run trend.
is what I meant to say.
There was a real mangling:
http://research.stlouisfed.org/fred2/graph/?chart_type=line&s1id=GDPC1&log_scales=Left
The g(t) in the discussion comes from fitting the graph to f(t) = a + b t/T + o(t^2/T^2) + g(t) where dg/dT ~ 0 and g = g(t,t0).
Tyler,
“In my view, supply-side factors are the main reason why the employment-to-population ratio has been so dismal since 2000, demand-side factors are the main reason why so many bad things have happened since 2007-2009, and supply-side factors and mismatched expectations are the fundamental reason why demand-side factors went south in 2007-2009.”
I am a bit on the fence when it comes to this. Supply-side is very very plausible. However, if you look at NGDP growth since 2000 or if you simply try to see whether all movements in V were offset by movements in M on a year-year basis, then you will see that the Fed has been passively tightening.
You can actually see a Philips-curve over that period, you’re not supposed to see that if monetary policy is doing its job properly. So even if supply-side factors play a role, monetary policy isn’t exactly helping.
But I’m not aware of anyone who thinks the long run does not matter; there are real live politicians who seem to think that pro-cyclical fiscal policy in the shrot run does not matter.
Mike takes the proper approach to the long run from micro economics.
So maybe a little modification for a macro approach is that the long run is as long as there is sufficient unemployment to make labor the abundant and cheap factor so business will continue to favor labor intensive rather than capital intensive means of production so productivity and living standards remain depressed.
A strict micro approach, as the basis for macro, would have to ignore, and thus not comprehend, irrational conduct by consumers who become irrationally exhuberant or pessimistic. So, when the herd goes in one direction, heading for the cliff, someone has to lead in the opposite direction, picking up the lost demand and assisting in the retraining of workers.
Or, head off the cliff. Your choice.
The issue with that Spencer is that labor isn’t as abundant as it may seem right now, even though the unemployment rate is 9%.
Yes, we understand the possibility of a fallacy of composition, but I think that hand is greatly overplayed. It’s one thing to opine the existence of a Paradox of Thrift and quite another to prove its existence and know exactly how we proceed to escape the trap.
Has anyone provided empirical evidence of a Paradox of Thrift other than suggestive two-variable coincidences?
I don’t believe there was anything “irrational” about the housing and financial bubble. To the contrary, I think economic actors were responding quite rationally to incentives. They may have been short sighted, but that’s not irrational. Many people’s expectations were justified ex post; only a relatively small number at the peak had ex post unreasonable expectations.
Keynes spoke much about pessimism, but again I don’t think pessimism is the cause nor is enthusiasm the solution. Deleveraging isn’t necessarily a result of pessimism. Businesses can squeeze out more life from existing capital without taking the bait of low interest rates.
Herding over a cliff is an overused and inappropriate analogy. 90% of the labor force is still working. There’s plenty of opportunity for those who want it. No one needs to “lead” anyone from their demise. This is the mindset of a serf.
I’m sure Bastiat would have something brilliant to say right now.
In order to take a strict micro approach, which relies on the rational actor, you would have to say this: “I don’t believe there was anything “irrational” about the housing and financial bubble.”
Think about that for a moment.
Have we learned anything about how rational the economic actor is since the 18th century when the classical model was postulated? I think we have.
When a glut of foreign capital is feeding financial markets, and our government is chanelling capital and credit to housing and financial markets through incentives, subsidies, and guarantees, decisions which are ultimately bad for all of us are perfectly rational for individuals.
Irrationality is not a necessary condition for bubbles.
Mike, As a micro enthusiast, and a government non-interventionist, I do not see your arguement and how you can deal with the inconsistencies.
Taking your points in order:
1. “When a glut of foreign capital is feeding financial markets,”–there cannot be a glut if a rational foreign capital seller, cognizant of all risks (including cognizance of a possible bubble), is channelling capital into our markets, or else there is no efficient market; or, if the foreign capital seller is channeling it into our markets for other reasons–supressing their own currency–then you are arguing for government intervention to stop it;
2. “Irrationality is not a necessary condition for bubbles.” That one is difficult to counter other than a statement that blue is green or red is yellow. Neither makes sense, and is contrary to the very definitions of the words. If you knew there was a bubble and its consequences, there would never be a bubble. I wouldn’t have purchased the tulip for $100 if I knew it was going to $2.
2.
Bill ain’t too bright, and doesn’t really understand what a bubble is. And also the fact that no bubbles exist.
Why isn’t there a better measure than core inflation? It seems like a cop out to say, “Well, energy and food are too volatile, so we’ll just strip them.” At least today. Really, there’s no way to create a model that, say, takes a three-month average?
Am I missing something with the employment-to-population ratio? The common narrative that “employment sucked during the housing boom too” seem to have a terribly short time horizon for employment-to-population ratios. Yes, the ratio declined sharply from a 1999-2000 peak of about 65%, but the peak of the 00′s boom had about the same ratio as the peak of the 80′s boom: 63%. The peaks before were far below that before the 80′s due to lower female participation, but wasn’t that effect pretty much over by the 63% in the late-80′s?
Basically the employment-to-population shows the same thing as the unemployed-to-labor-force ratio (i.e. the unemployment rate). Employment did not recover to “previous” levels in the late-90′s because the late-90′s had extraordinarily low unemployment. But employment DID recover to circa-1996 and the previous 1989 peak, just like unemployment recovered.
So putting aside the argument over what caused “lower employment” in the 00′s, we’re left only with causes over today’s high unemployment and low employment. And a traditional vicious cycle of lower demand and higher unemployment (due to sticky wages) is clearly the most empirically valid explanation for today’s unemployment. A structural unemployment cause makes absolutely no sense for, say, the massive amounts of teacher layoffs. Have so many teachers really become Zero Marginal Product workers overnight? Of course not. Then those laid-off young teachers go back to live at home, decide not to buy a new car, and so on.
And no, the government does not lower teacher’s wages across the board instead of layoffs. That just never happens to the extent necessary. Such a theoretical world would keep the cycle from happening and prices would go down in lockstep with the decline in teacher’s salaries. Instead, prices stay sticky (due to the stickiness of the main cost component in prices: wages) and real production goes down with a lot of excess capacity.
Yeah, zero interest rates and 600 billion of QE might not do enough to combat such as deflationary cycle. My reply to that is…then do monetary policy until it bloody does work. Simply doing price-level targeting instead of inflation “targeting” which really uses 2% as a ceiling rather than a target would be a huge improvement, and there’s no reason the Fed would have inability to inflate a currency at 2% instead of 1%. The argument over fiscal policy does pretty much nothing except obscure the fact that the Fed has the power to dramatically reduce unemployment and has chosen not to use that power.
If we start counting from 2000, there are only a few years left.
3.8% ain’t no joke, yo. It would be interesting if doctors only measured your “core” temperature and threw out the volatile part.
The short run and long run depend on how long it takes to adjust from the rigidity. This means its different for different types of crashes. For stock crashes the short run and long run can be days or months. For housing debt it can be 15 or 20 years as NGDP and debt loads adjust.
This also means that default should accelerate the proceses, and bailouts slow it down.
Point1: OK, fine, the ‘short run’ is not less than 5 years. Then, what do we call the relevant period for fiscal stimulus, QE, etc? The uber-short run? The ephemeral run? The myopic run?
Point2: If you accepted that 5< shortRun <15, it should be 1996-2006. In other words, no Obama or Eurozone prognostications.
Deal?
I would be happy to
All of those effects are speculated at less than 5 years
within 5 years time?
Semantic arguments are idiotic. You can have whatever terms you want, just make them relevant.
The answer is not TIPS, the answer is NGDP.
gamesliga bahis sitesi.
gamesliga yüksek oranlı iddaa oyna.
gamesliga gamesliga referans kodu servisi.
gamesliga gamesligaya üyelik üye olma
gamesliga referansı bahis sitesi.
gamesliga referans yüksek oranlı iddaa oyna.
gamesliga üyelik gamesliga referans kodu servisi.
gamesliga referans gamesligaya üyelik üye olma
Comments on this entry are closed.