Economics
Hysteresis
The cost of sustained unemployment
Feb 6th 2012, 20:46 by R.A. | WASHINGTON
I WAS going through some old open tabs in my browser this morning, and I came across this post by Brad DeLong, in which he walks through work on America's labour market that seems to be informing a forthcoming Brookings paper with Larry Summers. He takes us through an interesting exercise. Based on changes in America's labour-force participation rate and employment-population ratio over the past two years, Mr DeLong puts together an estimate of the impact of labour-market weakness on the long-run employment-population ratio. And based on that information, one can estimate the effect on potential output, which he suggests may have fallen by 0.5%:
If so, then the experience of the past two years provides enough information to produce a one-episode estimate of the labor-side hysteresis parameter η needed for the simple analytical framework. Two years during which real GDP has stayed flat at 7% below our pre-2007 estimates of potential output have managed to push potential output down by 0.5%: that suggests a value for η of 0.5/(7 x 2) = 0.035...
[A]nd with a value of 2.7%/year for the long-term growth rate g of the American economy, such a value for η looms very large in the social-welfare cost benefit analysis indeed. With a real social rate of time discount rd of 5%/year, η of 0.035 produces a present value of gross benefits from expansionary fiscal policy at the margin 2.5 times as large as simple multiplier calculations focused on current output. With a rate of time discount of 4%/year, it is not 2.5 but 3.7 times as large. And with a social rate of time discount of 3%/year, it is not 2.5 or 3.7 but rather 12.7 times as large.
The social discount rate of 4%/year also allows us to do a calculation of the cost of each extra month’s delay in the coming of a recovery proper to the U.S. labor market. Inserting an extra month with the output and employment gap at its current level costs the American economy roughly $100 billion in foregone immediate output. And, at a social real discount rate of 4%, if the reduction in labor-force attachment is indeed permanent, it also costs the American economy $270 billion in the present value of reduced future potential output.
That's a lot. And while there's a lot of uncertainty surrounding these estimates, the broader story—that sustained cyclical unemployment will raise the structural rate of unemployment, reducing labour supply and potential output—is both simple and compelling. Mr DeLong specifically mentions fiscal policy in the analysis above, but one can't help but think through these dynamics when looking at the Fed's economic projections, which note that—given what the FOMC considers to be an appropriate monetary policy—the unemployment rate is unlikely to be much below 7% a full 2 years from now.
In this blog, our correspondents consider the fluctuations in the world economy and the policies intended to produce more booms than busts. Adam Smith argued that in a free exchange both parties benefit, and this blog's aim is to encourage a free exchange of views on economic matters.
Advertisement
Money talks audio
- Analects | China
- Americas view | The Americas
- Babbage | Science and technology
- Bagehot's notebook | British politics
- Banyan | Asia
- Baobab | Africa
- Blighty | Britain
- Buttonwood's notebook | Financial markets
- Charlemagne's notebook | European politics
- Clausewitz | Defence, security and diplomacy
- Democracy in America | American politics
- Eastern approaches | Ex-communist Europe
- Elysée | France’s presidential election
- Free exchange | Economics
- Game theory | Sports
- Graphic detail | Charts, maps and infographics
- Gulliver | Business travel
- Johnson | Language
- Leviathan | Public policy
- Lexington's notebook | American politics
- Newsbook | News analysis
- Prospero | Books, arts and culture
- Schumpeter | Business and management
Most commented
- The euro crisis: The Greek run
- Foreigners in China: Barbarians at the gate, again
- Brazil’s economy: A bull diminished
- Russia and NATO: Rethink the reset
- Gay marriage: North Carolina begs the question too
- Nationalism online: Backfired
- Turkey’s cultural ambitions: Of marbles and men
- Greece and the euro: Exodus, chapter 1
- India-China relations and the media: Blame the messenger
- Politics and private equity: Bringing the Bain
Over the past five days
Most recommended
- The euro crisis: The Greek run
- Greek politics: Slouching towards the drachma
- Foreigners in China: Barbarians at the gate, again
- Brazil’s economy: A bull diminished
- The euro crisis: Going, going, gone?
- Jacob Zuma: Portrait of a president
- Rival versions of capitalism: The endangered public company
- Bo Xilai v Chen Guangcheng: Who is the mightier?
- Facebook goes public: Not top of the pops
- France and Germany: The Frangela show
Over the past seven days
Advertisement
Readers' comments
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
One point (which I didn't see in Mr DeLong's article, but I only had time for a quick scan) is the demographic changes in the workforce. In particular, by 2 years from now we will be seeing a significant number of baby-boomers hitting retirement age.
Given the age-relationship we have seen in layoffs, and in re-hiring since, those in my generation are disproportionately among the long-term unemployed. Which means that some allowance needs to be made for them simply going from the "long-term unemployed" to the "retired" category of the workforce. Personally, I am probably going to be moving from the ranks of the underemployed (working part time rather than full time, at least on paper) to retirement sometime in that interval. And I am far from alone.
How did the notion of calculating economic "potential" come to be taken seriously?
It involves making utterly questionable assumptions about an alternative universe in which the unemployed all have jobs, that those jobs have the same average productivity as the real jobs in this world, ignoring the role of technological advances in destroying jobs faster than they are created, and the fact that unemployment is much more severe amongst the low-skilled, and that we have no idea how to "fix" people's skills.
After rising by 8.8% from 2009 to 2010,
non-business bankruptcies are running at a pace
through the 3rd quarter of 2011 of
DOWN 10.3% Y/Y.
Regards
I'm a little less than convinced of the apparent accuracy of this when , if measured in the same way as it was in 1930, US unemployment is around 20%, heading for three times the original discussed figure.
If this hysteresis is an issue then it is already considerably bigger than discussed.
We then come to the bit of what government should "do" to invigorate the economy.
One very effective way of doing this was to , perversely, do less or even nothing at all!
There is track record for this, in the late 1940s a British civil servant was sent to Hong Kong to assist the recovery of the local economy.
In the most astonishing bit of prescience he realised that the local economy was doing just fine all by itself and the best thing he could do was.....nothing!
The result is that the Hong Kong gdp has multiplied six or seven times to comparable to even US levels per head.
Government spending is in the black and is 15% of gdp.
The rest of the "Tiger Economies" took this approach, and indeed the man himself, to their heart, they put statues of him up, which is a major factor in why your (and our) manufacturing has gone east.
The approach, not the statues that is...
"Governments generally achieve the opposite of what they set out to achieve" Sir John Harvey-Jones, former head of ICI.
If you think about it Greenspan and bernanke as agents of government have failed to create economic prosperity, indeed the opposite, and have managed to set the conditions to overthrow the Tunisian, Libyan Egyptian and the Syrian(pending) governments!
The way to do it is for the government to do less, tax less and let the people get on with it, everything else is little empire building.
Hysteresis is inherent in economics, or in any social study. It is of little use to specify the current state without understanding how it came to be. Human systems depend upon memory, and it is silly to imagine that identical states imply identical time evolution from those states.
This is why I consider comparisons of the recent unpleasantness with other post-WW2 recessions a waste of time.
Maybe yes, but with such a policy-package the social discount rate may be higher as it will hardly give motives to induce the economy to replace unproductive investment with productive investment while it will boost the aggregate demand.
Rather, such a policy-package will discourage the economy to do so, because the consensus will prevail on the supply-side that such a restructuring is unnecessary as the authorities are doing measures to grow investment in a sufficient manner to grow the GDP nicely. Who in the private sector would want to change their behaviour when they are being satisfied with making money out of the conventional portfolio?
But, if the economy hasn’t replaced unproductive investment with productive investment by the time the authorities come to think of it as about time to fight inflation after the phase of reflationary policy during which the real-wages level and thus the purchasing power or propensity to consume (as for consumption, refer to the sixth paragraph) of the households as an aggregate entity has been decreasing, the reaction of the economy to the exit strategy will be stronger pressures to unemployment by that much, when it is more costly for the economy to replace unproductive investment, which will have been pampered, with productive investment, which have been oppressed by unproductive investment due to the liquidity-preference by which the former will have been enjoying nice low interest rates and the latter frustrating high ones, than if with another long-run policy-package in which, as I have always been advocating, the authorities should actively finance the motives of replacing the former with the latter (not productive investment directly!) by, for one, changing the economy’s internal geographical structure against urban sprawl.
The authorities (i.e. actually, government in this case, not central bank, because it is about portfolio or capital stock, not about capital flow) shouldn’t, except some special cases, directly support what the economy is thinking of as productive investment, because if the authorities commit themselves to it means that they commit themselves to ‘uncertainty’ (not ‘risk’!) by that much.
If you are to take hysteresis into consideration, you need to take the above as well, because it is probable that employment with unproductive investment is not easily transferable to employment with productive investment at the aggregate level – due to the principle of hysteresis.
Now, the US households are known to consume a lot, but, viewed from the an-investment-dog-wags-its-saving-tail perspective, they are not really doing the consumption activities. When household-loans increase when they purchase goods, the household sector is not really consuming with the corresponding increment of the loans. By purchasing consumer-goods and -services on credit, they as an aggregate sector are increasing saving against both consumption and its total income, because when household-loans increase that way investment increases in the non-households sector, increasing saving in a corresponding manner. This Copernican change in the relation of saving and investment, which part of the Keynes’ revolution, should not be made light of.
Hence, with the policy-package in question – increasing the aggregate demand by the investment in a direct manner whether by monetary policy or fiscal policy – may well only make things costly.
By the way, DeLong’s chart is as funny as an elephant’s Chinese calligraphy is:
http://delong.typepad.com/.a/6a00e551f0800388340167616b5ddf970b-pi
This means that there are always more or less motives in the economy to replace unproductive investment with productive investment (i.e. the green general trend line) in whatever case, but the present, wrong policy package or its classical ‘a-saving-dog-wags-its-investment tail’ proposition has been a nuisance to those motives. That is why the chart looks as messy as a walrus’s Chinese calligraphy is if not chimp’s. (A chimp gets so excited with drawing materials that we often find it utterly impossible to find a theme or trend in its calligraphy while both an elephant and walrus keep their composure sufficiently to draw as ordered).
Since the focus here is the US in the midst of a 1vs99 % debate, perhaps it’s worth reflecting on the fact that that pay ratios between CEOs and average employees have increased by nearly 700% in the last 50 years. Why not skim-off empirically unproductive senior-staff salaries with a maximum wage to drive employment with existing capital.
The present economic problem lies in two points:
1. The propensity to save is very high
2. The liquidity-preference out of speculative motives is very high
The first point corresponds with flow of national income. In case the households purchase more goods and services on credit at a national money-income, it is not consumption but saving that increases, because investment then increases in the form of household loans. This is the very point few people understand. People tend to think that a saving dog wags its investment tail, but the right thinking is that an investment dog wags its saving tail.
The second point corresponds with stock of saving or portfolio of investment. In the above case, when the households purchase on credit, it is investment that increases thus saving in a corresponding manner at the macro-level. The non-household sector increases their claims to the household sector by as much as the given loans, and through the financial system the household sector increases their saving by that much. Then, the portfolio of investment changes corresponding to the loans.
Naturally, it is reasonable for the authorities to commit themselves to creating a framework that will both reduce the propensity to save and improve the liquidity-preference conditions.
Changes to tax system, which should be more progressive than at present in the truly effective terms (i.e. considering exemption, deduction and evasion), would reduce the propensity to save to the private non-household sector in the way the propensity to save to the public non-household sector (i.e. the government) is increased. The public finances would improve that way, and it would increasingly be advantageous to creating a framework to improve the liquidity-preference conditions through internally inefficient but externally efficient projects of public investment.
The troubles over growth, employment and 1% vs 99% will mitigate that way.
I agree, Fundy. If long-term unemployment erodes human capital, that doesn't seem so much worse in the long-term than building human capital for unsustainable positions.
Doug, that (Say's law) may be perfectly true in an imaginary static world, but then how would you find if I said market-mechanism (i.e. the private sector at least) was rather pro-cyclical, particularly on interest rates, that its dynamics in the real world could, by itself, not sufficiently encourage motives in the open market to replace unproductive investment with productive investment – at least within a meaning time frame to the living people in the real world? Refer to my above comment, which contains some heterodoxy.
I'd agree with you that the open market is pro-cyclical or, at least, cyclical. I agree with you that suppressing interests, at least by mental model, reduces the value of capital. I think part of the problem that I (and I think Fundy) are having with R.A.'s reasoning and Brad DeLong's reasoning and Scott Summers reasoning is that it seems kind of rhetorical. The syllogism seems to be:
1. Unemployment is bad (and long-term unemployment is worse) so we need (choose one of an even more aggressive FED/more stimulus.)
2. You question the aforementioned policies.
3. You think bad is good.
There are few fans of long-term unemployment. I am highly in favor of whatever gets us back to work. I think we need to see some admission that federal debt ultimately has a cost to growth if not now then down the road, that monetary easing has a cost in inflation if not now then down the road and that NGDP targeting could go terribly wrong as it essentially did in the late 70s. And that all these strategies may not work at all.
I'm happy to admit in return that leaving unemployment high has a cost to growth and sucks for those of us in it.
I tend to credit people more who will acknowledge both the costs, benefits and uncertainty of a favored strategy.
I agree. My frustration with mainstream econ is their assumption that depressions are random events caused by unpredictable shocks. There is enormous evidence that credit expansion causes unsustainable booms that crash. Gorton ("Slapped by the Invisible Hand") has huge amounts of data to show it. Why won't mainstream business cycle theory incorporate that knowledge?
If credit expansion causes unsustainable booms, and depressions are the result of unsustainable booms ending, then the damage to lives now and in the future that credit expansion causes should make us rethink using credit expansion as a policy tool. Federal debt has similar problems.
Wouldn't an ounce of prevention be better?
Doug, seemingly, you are missing the point, which I intended to stress most in my previous two comments: Opposition to the money-neutrality postulate.
You may find both Ryan and Fundy taking the postulate into account in the opposite manners when they think of long-run monetary policy:
Ryan thinks that as money is neutral (at least in effect) an active and aggressive monetary policy will keep providing liquidity to the open market over the period of its process of replacing unproductive investment with productive investment (by what he may be thinking of as the market-mechanism’s autonomous function of portfolio restructuring) and thus the increment of the aggregate demand will eventually be filled with the increment of the materialised aggregate supply out of productive investment replacing unproductive investment.
(I think that such a market function do exist, particularly represented by the description of innovation, but at the same time that not always in a sufficient way to make productive investment outrun unproductive investment within the increment of the effective demand. In case of the function being insufficient, an expansionary monetary policy will materialise Fundy’s scenario, as I am explaining below).
Fundy thinks that as money is neutral expansionary monetary policy is not only useless for the process of productive investment autonomously (i.e. because of the money-neutrality postulate) replacing productive investment but also harmful to the economy as the policy will exacerbate the present situation in which unproductive investment is being dominant over productive investment. Unproductive investment is not marginal-efficient enough to give returns that should outrun the interest rate over the whole life of the corresponding capital-asset when investment grows, and hence the increased aggregate investment by an expansionary monetary policy will someday cause a storm of insolvencies. He thinks that the process of productive investment replacing unproductive investment will be hindered by an expansionary monetary policy.
A paraphrase of your missing point, therefore, is the perspective from which money or saving should be viewed as stock or portfolio (and not only as flow).
On the contrary, opposing the money-neutrality postulate, I believe that the process of replacing productive investment with unproductive investment can be financed in the way that the government directly induces and finances the process by using its portfolio function while the central bank finances both the government and, through the banking system, the process at the same time. In that case, an aggressive expansionary monetary policy like QE is unnecessary for long-run policymaking. (Such an aggressive policy cannot replace unproductive investment with productive investment to the sufficient extent while it is boosting the aggregate demand, and hence should be best used for short-run policymaking in case that financial-market panic is imminent or already panic-stricken, because it still can conduct its power to boost the aggregate demand). Then, try and read my first comment again and consider why I say “the authorities should actively finance the motives of replacing the former with the latter (not productive investment directly!)”
“if the reduction in labor-force attachment is indeed permanent, it also costs the American economy $270 billion in the present value of reduced future potential output.”
This is sad and serious. Don’t you think some effort ought to go into diagnosis instead of just fixating on prescriptions? If you don’t know what causes the disease, how do you know the prescription works? It could be causing more harm than good.
I find it hard to believe that “crap happens”, satisfies the majority of mainstream economists as an explanation for depressions. Isn’t it time to look at other explanations?
My diagnosis is that your chronic back pain was caused by a sustained period of back pain in previous years. Had you taken more painkillers then, you wouldn't be suffering now.