Tuesday, June 24, 2014

Summers on Stagnation

Larry Summers has published a very interesting speech, U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound. I heard a version of the same thoughts last October, at the joint Brookings-Hoover conference "The U.S. Financial System—Five Years After the Crisis."

I was struck then, as I am now, at how much consensus there is among macroeconomists. Yes, you heard it here. And Larry expresses it elegantly, as you might expect. While the press talks about recovery, macroeconomists look at output growth and employment and it still looks pretty dismal.

Source: Larry Summers
GDP fell in the recession, but has not recovered relative to trend. What was a recession is turning into everyone's nightmare, perpetually slow growth.

Usually, GDP rises back to "potential" or "trend." This time, the "potential" is falling to meet the lackluster results. Potential decline, Larry points out, already closed 5% of the gap, leaving only 5% left. This is starting to look like lack-of-growth theory, not business cycle dynamics.

The unemployment rate is declining, but the employment-population ratio has not budged, even looking at prime-age men to offset some of the demographic effects.

Source: Larry Summers
You have heard these very points on this blog and John Taylor's blog. Ed Lazear's slide deck starts the same way. Bob Hall's macro annual paper starts the same way. Charlie Plosser's speech starts the same way.  I repeat here, with Summers' pictures, to emphasize the extroardinary consensus.

It's also extraordinary because the Washington policy machine has gotten bored with growth or lack of growth and moved on to squabble about other things.

Now, to analysis. Larry makes a very interesting case that fundamental forces in the economy push us to low real interest rates. The ones I found most interesting,
First, reductions in demand for debt-financed investment. ...probably to a greater extent, it is a reflection of the changing character of productive economic activity... Ponder the fact that it used to require tens of millions of dollars to start a significant new venture, and significant new ventures today are seeded with hundreds of thousands of dollars. All of this means reduced demand for investment, with consequences for equilibrium levels of interest rates.
[Fourth] is a substantial shift in the relative price of capital goods [falling about 20% since 1980].. Something similar, but less dramatic, is present in the data on consumer durables. To take just one example, during a period in which median wages have been stagnant over the last 30 years, median wages in terms of automobiles have almost doubled..
In sum, one aspect of the new "stuff cheap, people [with skills] expensive" economy is a reduction in the real rate of interest. It took a lot of money to build railroads. It doesn't take a lot of money to build apps.

Source: Larry Summers
Summers points to an interesting calculation of the "natural" rate of interest by Laubach and Williams, which I will have to look up. So far, the "negative natural rate" at the basis of all the new-Keyensian analysis I have read has been a deus-ex-machina, not independently measured. Interesting.

Here, as must be the case, inevitably, we part company. It's a quantitative problem. The natural rate is per Laubach and Williams, about -0.5%. But we still have 2% inflation, so the actual real interest rate is -1.5%, well below -0.5%. With 2% inflation, we need something like a 4-5% negative "natural rate" to cause a serious zero bound problem.  While Summers' discussion points to low interest rates, it is awfully hard to get any sensible economic model that has a sharply negative long run real rate

Moreover, to Summers, the one and only problem worth mentioning in the US economy is that the "natural rate" is negative while nominal rates cannot fall below zero. Can't we think of one single solitary additional distortion in the American economy?

What to do? Summers sees the problem as eternal lack of "demand" and recommends more of it. I think this more of a microeconomic/lack of growth theory problem needing the removal of distortions.

We still agree a bit -- Summers starts with "There is surely scope in today’s United States for regulatory and tax reforms that would promote private investment." That's distortion removal.

And I am interested that his calls for stimulus do not rely on the idea that consumers ignore the fact that government debt must be repaid. "Although it should be clear from what I am saying that I do not regard a prompt reduction in the federal budget deficit as a high order priority for the nation, I would be the first to agree with Michael Peterson and his colleagues at the Peter G. Peterson Foundation that  credible long-term commitments would be a contributor to confidence." Apparently the confidence fairy reads Ricardo, as do fully-Ricardian new-Keynesian models.

It's hard to object to "policies that are successful in promoting exports, whether through trade agreements, relaxation of export controls, promotion of U.S. exports, or resistance to the mercantilist practices of other nations when they are pursued, offer the prospect of increasing demand," though that can quickly be misread as invitation to our own mercantilist efforts.

But this is all small potatoes, compared to 5% loss of potential, 5% loss of GDP, and a creepingly slow escalator, no? So, really the core of Larry's prescription is
as I’ve emphasized in the past, public investments have a potentially substantial role to play...ask if anyone is proud of Kennedy Airport, and then to ask how it is possible that a moment when the long-term interest rate in a currency we print is below 3 percent and the construction unemployment rate approaches double digits is not the right moment to increase public investment in general—and perhaps to repair Kennedy Airport in particular
It's hard to argue with fixing potholes, especially in Chicago. And there is no argument against investment that earns a positive rate of return. The question, though, is not whether those are good investments but whether making such investments can raise GDP 5%, potential GDP by 5%, and raise the desultory growth rate.  Even at a multiplier of one, there are not $750 billion of positive net present value roads and bridges to build -- and we haven't started with the opposition of anti-sprawl and environmental lobbies who don't want roads and bridges built in the first place. The Keystone pipeline, and LNG export terminals are infrastructure too. (I should be careful however. There surely are $750 billion a year of alternative-energy boondoggles to build!)

Here, of course, Summers thinks multipliers -- even multipliers for tax financed (notice the Ricardian comment) and wasted (that's in the models, the usefulness of the infrastructure has nothing to do with the multiplier) is in the range of 4 to 5.

To me, this is just magical thinking -- that the key to long run prosperity is government spending, even if wasted.

But these are old arguments, and I did not write to rehash old controversies. From my point of view, the speech is an eloquent statement of the problem, and the gulf between Summers and people who think like I do is much narrower than the gulf between macroeconomists and the policy establishment which is not even thinking about slow growth anymore. From my point of view, the focus on and evident emptiness of the "demand" solution -- its reliance on magic -- just emphasizes where the real hard problems are.

A last note of praise. Notice Summers said nothing about the minimum wage, the earnings of the top 1/10 of a percent, and other fixations of the current partisan squabble.


  1. " the actual real interest rate is -1.5%, well above -0.5%"

    Did you mean +1.5%?

  2. Oh Professor.... I love your optimism, I wish I had it.

    You say that there is a lot of consensus in the macro profession. I do hate to throw a bucket of very ice cold Lake Michigan water on that, but it seems to me that the consensus that you refer to is regarding the historical data. Larry Summers is very, very smart, and he and everybody else can see that actual GDP is simply not bouncing back to potential. This is not a hard consensus to form.
    It seems to me that your claim of the consensus starts breaking down after your question of "What to do?" It does not seem to me that you describe a lot of consensus there. Of course that among reasonable economists there will be some overlap in the normative world, but consensus? Larry Summers is a plain old fashioned Keynesian guy who believes in the Keynesian multiplier, and wants these - what was Summers' famous phrase? - "temporary, targeted" whatever stimuli programs.
    You are a supply side guy - you want to remove distortions, remove incentive-killing regulations, all so that the private sector can start investing, create productive jobs, etc etc.
    You also believe in rules and not in discretion, so that policy can be predictable by the economic agents. Summers is very, very far from being a "rules" guy. He always likes to keep his options open (which means, discretion).

    So, sorry Prof, I do not see much consensus here. But again, this is Manfred The Cynic.

  3. On the "natural rate of interest"


    "Thus, the natural rate of interest is the real fed funds rate consistent with stable inflation absent shocks to demand and supply."

    Let's see what is wrong with this statement:

    1. Nominal Interest Rate - The federal reserve sets a nominal interest rate for money to be borrowed at (discount window). The inflation rate (the effect of borrowed money) is realized after the fact. Right away the whole notion of the Fed setting a "real fed funds rate" runs into problems.

    2. Effective Interest Rate - The fed can lead a horse to water (offer money at a nominal interest rate), but can't make him drink (take on a loan). And so the whole notion that a central bank can set an effective nominal interest rate (people are willing to borrow at that rate) seems dubious.

    3. Establishing Causality - If changes in the fed funds rate can counteract a supply / demand shock then they can also cause a supply / demand shock. And so, what fed funds rate is consistent with stable inflation when the fed causes supply / demand shocks by altering the fed funds rate?

  4. Everyone always picks on Kennedy Airport. The JetBlue Terminal is brand new and really quite lovely. The real problem with the airport is the time it takes to commute to and from Manhattan. That's why many often fly into the easily more dated and deprecated LaGuardia Airport.

  5. Mr. Cochrane,

    You identified Summer's problem in your statement above (pardon my rewrite):
    === ===
    Here, of course, Summers thinks wealth produced is in the range of $400 to $500 for each $100 spent by government. He thinks this even for spending financed by higher taxes and even when wasted. The usefulness of any infrastructure built has nothing to do with the multiplier in his model.
    === ===

    This is (wild-eyed Keynesian) magical thinking as you correctly call it. Keynes famously called for reviving an economy by digging ditches and refilling them, using as many workers as possible. I think of it as the Bunny Theory of Money. If you send enough Bunny Money into the world, it will combine with itself and produce Bunny Money offspring, about 4 for for every adult Bunny Money.

    Some simple equations seem to say this, if one misinterprets them and reverses cause and effect. But, to actually believe this requires keeping a narrow mind buried well into a book and not looking around at reality.

    Once a person, even Summers, spouts magical thinking on subjects deeply within his area of supposed expertise, why should I believe anything he says? I don't care how erudite Summers is in setting down all of the problems, questions, and strange facts facing us. I can have no confidence in his prooposed solutions.

    It is an old trick to seem wise by reporting the news in detail. TV anchors do it all the time, and it why many people want to be a TV anchor or reporter. They know and report many facts which are absolutely true, then ... so what?

    The commonly taught and usually believed Keynes Multiplier is probably the most destructive piece of false economics. It provides support for government borrowing and spending regardless of the merits of the project. The act of spending is supposed to create more wealth than the wealth wasted by the spending. So, why not spend big?

    But, the commonly taught derivation is laughably illogical, not merely a bad estimate.

    The Illogic of the Keynes Multiplier

  6. I would recommend reading a recent Bain & Co. report "A World Awash in Money." They say there is a "superabundance" of capital globally. In short, huge pools of savings or capital are being generated in semi- or non-market situations.

    Okay, so who knows what should be "real" interest rates (in current context), or if our gauges of inflation are accurate?

    Maybe the PCE overstates inflation by 2 percent (some argue so, and point to today's incredible gadgetry and medicine, simply not relatable to the even as recent as the 1980s).

    With current supply and demand, maybe the "real" rate of interest "should" be negative three or four percent, but we hit the zero bound, and government intervention in the form of deposit insurance and TIPs etc.

    BTW, if there is a glut of capital, then savers should expect losses. That is, in free markets, savers are entitled to returns---and to losses. If the amount of savings exceeds good places to invest, some investing has to not pan out. In aggregate, savers "should" lose.

    If central banks target low rates of inflation, then we never see the real negative rates of interest we need to see. That is, we need negative four percent interest, but the central bank hits 1 percent inflation, and capital markets hit ZLB. No signal is sent to savers to stop saving. And anyway, the sovereign wealth funds and many pension plans are on automatic.

    One answer is way more federal borrowing and spending. But I also worry about who pays it back, and corruption in spending.

    So, I like huge tax cuts for the middle class, financed by QE. The Fed buys a trillion in Treasuries, and we give a trillion tax cut to the middle class, maybe FICA cuts. Or (blasphemy coming) the Fed prints a trillion and gives it to Social Security-Medicare, and they cut taxes for next two years.

    I also ask that Cochrane re-cast his recent Hoover presentation here, the one where we convert federal debt into bank reserves. This ties in nicely, no?

  7. John,

    Since 2008 we all also started to have smartphones.

    - With GPS in our hands, we reduce our petrol usage by finding the fastest route and not getting lost.
    - The collaboration economy like AirBnb accounts as $20 per night instead of $150 in a hotel by making a better utilisation of assets.
    - Others like Uber simply make the acquisition of cars almost irrelevant, reducing GDP but significantly increasing the purchasing power, sustainability and lives of people, simply because there is less waste as assets like a car are not used 3% of the time buy 95%.
    - We don´t need to buy so many cameras because our phones are good enough
    - We don´t need to spend on phone calls because Skype delivers voice free
    - Comparing sites find you better deals for insurance or mortgages.

    All of these bring prices and GDP down, but it makes us all "wealthier", which is good.

    This might account for a small or large part of the problem, I have not run the numbers, but don´t you think that macroeconomists should start to think about real economics (value created to society) and not accounting (GDP?).


    1. Technological improvements dont reduce GDP
      Read the fallacy of the broken window by Bastiat, you can easily find it online

  8. >> "There is surely scope in today’s United States for regulatory and tax reforms that would promote private investment." That's distortion removal.

    Not after Congress finishes auctioning favors to K Street. It would be reform in the sense of our recent health care reform - an orgy of special interest spending resulting in disaster.

  9. He wants to have it both ways!
    More govt. spending, but he is concerned about deficits.
    And why on earth do airports need to be financed with taxes?

    1. "And why on earth do airports need to be financed with taxes?"

      Perhaps because there is no way any business would ever invest in an airport without huge government guarantees and subsidies. Transportation businesses can never make money over the long haul and airports are a system cost that will never be profitable for any length of time. If we want transportation, we need government spending and taxes to fund it. You can go back 5,000 years, but that's the story from rent-a-mule, to the post-chaise, to the railroads, to the airlines, to the Star Trek Express to Rigel V.

    2. "Transportation businesses can never make money over the long haul"
      Who said that? Moses?
      Airports can perfectly be profitable and financed with fees paid by the passengers. It works the same way than a shoe store. If you don't agree then you don't agree with free market capitalism and the price system to allocate resources.
      Also, why should a poor person who never uses airplanes pay for Kennedy airport expansion?

  10. "Summers sees the problem as eternal lack of "demand" and recommends more of it. I think this more of a microeconomic/lack of growth theory problem needing the removal of distortions."

    You should write a post expanding on this. Presumably most of the distortions were already in place in 2007 and cannot explain the slow growth today. Are you suggesting that regulations and policies put in place after the crisis (dodd-frank, stimulus, etc.) are responsible for the slow growth? or that the distortions in place before the crisis (cra, moral hazard, etc.) led to the crisis and the slow growth?

    1. Dodd Frank, ACA, etc. are worse. But no, they didn't cause the recession as you point out. I think they, and social program disincentives, make the economy much less resilient to shocks. Take social security disability, 9 million and climbing. When people have jobs, it's not that relevant. If they lose jobs in the recession, go on disability, and then never get out, that's a permanent loss.

    2. But, they are not people who want free checks, they are really disabled people who found themselves in a situation where they had no other way out. Reducing their right to disability without creating more jobs would do nothing for economy and would increase their suffering. And creating enough jobs so that they can continue working would make any cuts in disability rights unnecessary, because they have shown that the prefer to work.

  11. It's time to redefine the recovery to something more accurate than jobs and GDP. Those two indicators are only small blemishes of part of an otherwise rosier picture. As far as Wall St. and profits and earnings are concerned, there's no stagnation.

    1. If only we all lived on Wall Street instead of Main Street.

      This reminds me of William Dudley's infamous "Let them eat iPads" speech. I think you can expect a job offer from the Department of Hedonic Adjustments pretty soon.

      Earnings are easily manipulated numbers. You borrow at repressed rates, buy back shares and voila! Headline EPS goes up. But this borrowing does nothing to add production capacity or start new projects. It doesn't add new jobs. It’s money spent for the sole purpose of manipulating a number. To make matters worse, it's done by buying back shares while they are making all-time highs, which from a business POV is insane.

      Then they play the "non-GAAP/pro forma earnings/one time write off" game and hype future earnings. And then those earnings beat expectations because as earnings season approaches the estimates start falling.

      Ever wonder how S&P 500 revenue growth from March 2009 to March 2014 averaged just 3.2% while earnings growth averaged 16.2? That's a pretty amazing jump in efficiency! And some of it really was increased efficiency, like getting more work out of the employees who survived the layoffs.

      You can only play that game for so long. The math always rises to the surface, like a bloated corpse in a murder mystery.

  12. I wonder if the success of policies favoring wealth accumulation and concentration since 1980 haven't created an ecosystem where it is much more profitable and far less risky to simply use financial maneuvering, rent seeking and government support to enhance wealth, making investment in a physical economy or productive enterprises appear far more risky and difficult and therefore unattractive to capital owners?
    Wouldn't confiscatory tax regimes make investment in real economy sectors more attractive? Should there be a tax regime in place that encourages capital owners to "get off the pot" and put that capital to work for the benefit of more people?

    1. "Wouldn't confiscatory tax regimes make investment in real economy sectors more attractive?"

      The problem with confiscatory tax regimes is that they tend to throw the baby out with the bath water - meaning in the spirit of "fairness" tax regimes are primarily tilted towards flat tax, consumption tax, or the ever popular redistribution of wealth - no matter how you obtained that wealth.

      "Should there be a tax regime in place that encourages capital owners to get off the pot and put that capital to work for the benefit of more people?"

      Yes there should, but what you really want is risk inducing tax policy not confiscatory tax policy - not the same thing.

      Psychologically - people are more inclined to take positive action in response to positive motivation (take risk / receive reward) rather than negative motivation (take no risk / have property stripped away).

      Risk inducing tax policy would require the federal government to sell a risky asset with a rate of return that is realizable against a future tax liability - meaning term structure tax breaks that are bought from the federal government. Purchaser of tax break gives up liquidity for potential return on investment.

    2. Changes in tax policy won't address the fact that a little money spent buying favors from the government has a much better ROI than a lot of money spent on capex.

    3. Yes, extremely high tax rates would help. The government could then hire lots of people to dig ditches and fill them in, or perhaps to update their Facebook pages every five minutes. These people would then have money and that would encourage businesses to invest in order to provide goods and services that these people might give them money for. It has worked in the past.

      In the 1950s and 1960s business magazines used to have pages of income statistics broken down regionally. Businesses would then look for places where people had money and invest accordingly. The businesses got money. The businessmen got rich. The people got the stuff they spent their money on. Then came the tax cuts and the assault on "wasteful" government spending and businessmen whining and whining. If people had more money I'd invest in an earplug factory.

  13. Since Andrew Garland already emphasized so well the destructiveness and sheer idiocy of the Keynes Multiplier, and the consensus amongst professional economists is only on the problem not the solution, I'll once gain devote myself to beating the dead horse of my "Existential" (in the 50's French philosophy sense) economic theory (which I have been unsuccessfully flogging for years) back to life.

    Let me try explaining it this way: It's possible (for a period of time) to have a country in which almost nobody works. All manufactured goods are imported, construction crews (and material) for bridges and other infrastructure are imported, and what few jobs are available are mostly low-paying food-service jobs.

    How is this possible? After all, knowlegeable (that is, free-market) economists tell us that dollars spent buying products from overseas have to come back, and when they do they will create other jobs. True... unless those dollars are used NOT to buy products or services from us, but to buy assets we inherited from previous generations (buildings, companies, etc) or nature (forests, land, et cetera).

    I contend this is exactly what has been happening in the United States for the last several decades. And that by comparison most of the things professional economists focus on is pretty trivial. And I fear that unless professional economists embrace this theory the United States is doomed.

    (P.S. I call it my "Existential" theory because the old existentialist philosophers argued that by default most people will gladly give up their freedom in exchange for various [perceived] benefits.)

    1. The problem with Garland et al is that Keynesian multiplier still seems to work like a honey. Economists can jump up and down and argue that it doesn't exist, but every time the government pisses off money on poor or middle class people, the money winds up getting multiplied. When they give it to billionaires, they might as well have burned it, except that the fire at least that could have heated a room or something.

    2. To Neal Reynolds,

      Thanks for the shout out.

      -- --
      To Kaleberg,

      Why do you think that? A citation to an analysis of observed data would be interesting. What do you literally mean by "the money gets multiplied"?

  14. It is not a fact say that government debt must be repaid. Debt burden can shrink as a percentage of the budget while debt increases in absolute terms. That's simple math.

    1. But the prsent value of primary surpluses still equals the outstainding debt in that circumstance. That also is simple math. r>g means there is a government budget constraint.

    2. "r>g means there is a government budget constraint"

      Not really. The burden of government debt is the interest only. A government can roll over the principle of it's debt indefinitely - it never "has to" be repaid. The interest is paid from tax revenue (assuming no Ponzi finance).

      Assume that tax revenue is a constant percentage of nominal GDP. And so at a glance it would appear that annual nominal GDP must be greater than the average interest rate times the total debt outstanding otherwise interest payments as a percentage of tax revenue grow toward and eventually exceed 100%.

      BUT, the underlying assumption is that the federal government makes regular cash interest payments on its debt (coupon bonds). This is not the case with some forms of government debt. In some instances, the federal government sells bonds where the interest accrues - interest and principle are returned when the bond reaches maturity. And so a government when faced with chronic budget deficits can switch to more accrual bonds and lengthen maturity.

      What is the interest rate on 100 year government debt? How about 500 year?

    3. If a government runs a 5% deficit and the term premium is 0.25% added interest for every 1% increase in duration then the government can extend duration by:

      5% + 0.25% x ( X / 1% ) = X

      X = 6.66%

      And pay the same amount of annual interest. In essence the government is stretching out the interest payments over more years.

    4. England used to sell consuls, debt with no due date. I think they paid 3%, or at least that's what Thackeray said in Becky Sharpe. If nothing else, Thackeray knew his money.

    5. Kaleberg,



      They started out at 3%, and the interest rate was reduced over time. Also, British consols were coupon securities that make quarterly interest payments. I was referring to bonds such as:


  15. "While the press talks about recovery, macroeconomists look at output growth and employment and it still looks pretty dismal."

    Looking at the GDP growth graph, I don't see that. Growth is fine. The problem is that we never got the bounceback special growth that usually happens after recessions. Or, phrased differently, potential GDP fell in 2009, we recovered in 2010, and since then have grown at a normal rate. My interpretation is that Obama Administration policies created a huge one-time loss for the economy but did not affect the rate of growth. A smaller fraction of the population is employed than in the old regime, and that's not going to change, but growth will be at a normal rate.

    1. Most of the loss is between 2008Q1 and 2009 Q1. Obama took the office in 2009 Q1. There is no way he could do the damage. If the loss is the result of wrong policies, it could only be pre-Obama policies. Effects from Obama's policies would be visible late 2009 at earliest, and 2010 more likely, the time of recovery. Although, growth is constantly lower than the prediction of 2007 potential.

    2. Let me clarify. It's *potential* GDP that fell in 2009 (and after, I suppose, as Obama implemented more policies) because of Obama. Actual GDP had already fallen, but if potential GDP had not fallen, actual GDP would have bounced back. Initial reduced employment was because of the recession, but then extended unemployment benefits, means-tested mortgage relief, the prospect of Obamacare, easy disability benefits etc. prevented employment from increasing as much as it woudl have ordinarily. Structural damage like that to an economy, however, is a one-time shock, and needn't be a drag on growth,even though the annual losses it causes persist forever.

    3. Extended unemployment benefits and disability benefits cannot influence the number of workers in the economy where number of unemployed is more than order of magnitude greater than number of working places. The problem was number of available jobs, not how much incentive people have to seek new jobs. Even if you believe that 50% of people on benefits seek less (way, way, too high percentage), that still leaves at least 5 unemployed for every new job. Meaning, at least 80% of those willing to do anything for a job still could not find one, instead of 90%.
      Unemployment benefits are not something you live on, it's something to help you not to starve while trying to find a new job. If economy is close to the full employment, then you can say that small amount of people willing to live on dole may matter. As long as unemployment is high, economy won't notice them.
      Also, if you look at the GDP and unemployment figures, job recovery has been much slower than GDP recovery after recessions at least since 70s.
      When benefits run out, you see reduction in unemployment rate because some people simply lose hope and stop looking for job. That is structural damage to the economy.
      Another structural damage to the economy are long-term unemployed. Analysis seem to say that people who have been unemployed more than a year have additional trouble finding jobs, because employers think there is something wrong with them (even if the only "wrong" thing was that 10M people were downsized when only 1M jobs total could be found throughout a year). When they find one, some have difficult time returning to work habits.
      Look at employment/population ratio for men 25-54. It dropped almost 7%, and it has now recovered around 2%. It's hard to imagine that that many men in prime will stop working because of (meager) benefits.

      When crisis strikes, economy will, sooner or later, recover. The questions are: how soon will it recover, how much permanent damage will be left, and how severe will the impact be on the people.
      Without any government intervention, impact on those with lower income (those who can't build themselves emergency fund) can be disastrous. Government should ensure that they have some basic needs (food, place to sleep), if nothing else, to keep them alive and capable of rejoining workforce later. As for the economy...?

    4. For big recession, I would actually like to see full Keynesian experiment. Pick a sector. Many crises are hard on construction sector (when contracting, businesses do not need new buildings, people do not buy new houses, nobody needs new infrastructure), so there should be lots of unemployed resources. Government should create, quickly (inside 3, at most 6 months) enough construction projects to keep construction industry fully employed for at least a year, maybe two. Government should run as high deficits as needed (no contraction elsewhere to finance this). Possible outcomes:
      (1) Keynes was right, and full hysteresis multiplier exists (around 5) - excellent, economy is booming again, or will boom shortly
      (2) Keynes was right, but with lower multiplier (around 1.5) - also great, recession can be dealt with quickly, spending from construction sector has propped up other sectors
      (3) Utter fail, lots of money spent for no effect on GDP - actually not very bad result - because of the failure and all the new debt, USD would devalue, probably even against China and India; there would be reorientation on the markets (US products would replace imported ones), exports would be much cheaper, and some industries would even return to US - economy would boom soon
      (4) Middling results, economy would have some positive results while the program is going, but that would fade after the construction cycle is finished, without lasting effect on the recession and economy - really bad result, basically prolonging the recession, while mitigating early impact

      As long as cost of (4) is not prohibitively expensive compared to the loss in the economy as a result of the current crisis, that would be worthwhile experiment. It can hardly be worse than the response to this crisis.

      (Additionally, if Keynes was right, government would not need enough worthwhile projects to fully employ construction industry - even if return of every single project individually is below 1, employment of people who would otherwise be unemployed and resources that would otherwise be idle would generate taxes, additional demand, jump-start other sectors and industries, generate additional taxes and demand to the point where private sector would be able to create enough demand for construction industry as government projects end. At that point any actual value received from those projects would be bonus.)

  16. You say that "people with skills.. expensive." But isn't the high unemployment rate for the unskilled an indication that all labor is expensive? Low skilled workers are expensive because their value relative to the costs (both direct and indirect) of hiring them is too high. The price at which they would be valuable would not only be below the minimum wage but would require removing some of the regulations on labor, on health care, hiring and firing, etc. Absent such changes, their true costs are too high.

  17. Great post as usual. On your statement "while Summers' discussion points to low interest rates, it is awfully hard to get any sensible economic model that has a sharply negative long run real rate", I was at an LSE seminar a couple of days ago where Gautti Eggertsson presented a paper trying to do just that:

    I wonder if you have any comments on that model.

    1. Juan,

      From the link:

      Page 27: "We can combine (7) and (38) to obtain a no arbitrage condition linking the capital rental rate to the real interest rate"

      No you can't for several reasons. Most notably market segmentation may allow for arbitrage opportunity.

      For instance the U. S. central bank is limited by law as follows:


      "Notwithstanding any other provision of this chapter, any bonds, notes, or other obligations which are direct obligations of the United States or which are fully guaranteed by the United States as to the principal and interest may be bought and sold without regard to maturities but only in the open market."

      As such, the capital rental rate may exceed the real interest rate precisely because the central bank is limited in what it may purchase.

      Page 27: "Relative to our earlier derivation of loan supply in (10), we see that the last term on the right hand side is new - the presence of capital will reduce the supply of savings available in the bond market. However, as before there is no reason for the two curves, Lst and Ldt , to intersect at a positive real interest rate."

      They will if you allow for an arbitrage condition between the rental rate on capital and the real interest rate.

  18. "Ponder the fact that it used to require tens of millions of dollars to start a significant new venture, and significant new ventures today are seeded with hundreds of thousands of dollars."

    This is actually a good thing, right? Financial capital is a scarce resource, so if new ventures require less capital, then new venture formation will be less constrained, which should be a positive for economic growth, right? The effect would be similar to a case where new technology allows us to produce more output with a fixed amount of energy. Understood that equilibrium interest rates will be lower, just as new energy-efficient technology will lead to lower energy prices, but the impact on growth should be positive, correct?


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.