September 2022

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  
Blog powered by Typepad

« An Open Challenge to Rachel Maddow | Main | Going to Have to Face It You're Addicted to Debt »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Today is Lucille Ball's 100th birthday. So distract yourself.

Who says the stimulus failed? You? It wasn't just big enough to make a real difference. Relax and go back to your cave.

To connect posts, the Hoover Administration was the first to take an activist response to an economic downturn. It turned a conventional downturn into the Great Depression. Bungled Fed policymaking contributed, of course.

In all prior donwturns (including the severe contraction of 1920-21), the response was to let markets adjust. And they did, as convetional economic theory predicts. Of course, 1920-21 event was the first in which the US had a central bank. The Fed did little until the depression was almost over. Meltzer attributes recovery to the real-balance effect. Deflation is its own cure.

It is useful to reiterate that the Austrian monetary analysis of economic fluctuations is just basically the classical economic view. In the 19th century, it was stated early and clearly by Henry Thornton.

Steve, obviously I agree with your sentiments, but I could see an interventionist say, "What do you mean, TARP failed?" They claim the gov't made money on it, it resuscitated asset markets, etc. It's not obviously wrong on the face of it. I.e. I think it's wrong, but only because I'm relying on Austrian theory to speculate on what the world would look like if they hadn't done TARP.

In contrast, I think it's clear that the stimulus quite obviously failed.

Thank you Stephan for making my point better than I could have. And we could do without the nastiness too.

I disagree, QE1 and QE2 worked. American companies borrowed money (at low interest rates)and with it purchased equipment manufactured in the BRICs or less developed economies or built factories in those countries where labor costs are a fraction of US labor costs, health care and other employee benefits barely exist and labor standards look like they were drafted from a Dickens novel. (Tax differentials is the biggest red herring ever manufactured in this country for losing its competitiveness.) The result has been healthy corporate profits.

Undoubtedly, I will be labeled by many who read this as some liberal unionist, but I'm sincerely not. What I am is a realist, someone who has personally engaged in outsourcing activities fully aware that such a decision would put US employees out of work. But the economics were so compelling that deciding otherwise would have been stupid for the enterprise.

Until we come to terms with the true impact globalism has had and will continue to have on our economy, our economic woes will surely continue.

Of course, until the legal profession becomes as widely outsourced (some legal outsourcing currently exists) as manufacturing and many service jobs have become, nothing will change. Because lawyers write the laws, they will surely craft legislation to prevent the migration of their jobs and hopefully bring back many in other industries that have left our borders.

The challenge will be to do this without triggering global trade wars.

As Jerry says, flexible wages and prices and the resulting increase in the real value of cash balances was an essential ingredient in the correction and "re-balancing" from the 1920-1921 downturn of the American economy following the war and post-war booms.

Of course, this argument might also be used in response to some in our own Austrian camp who argue that "MV" should be kept constant in the face of an economic downturn in which there would otherwise be downward pressures on prices and wages in general.

And if I may add one of my usual "doctrinal" notes, the modern "real balance effect" is also often called the "Pigou Effect," since Arthur C. Pigou published a couple of articles in the 1940s explaining the influence on desired cash balance holding due to falling prices as a logical self-correction to falling "aggregate demand."

But actually, this "modern" real cash balance correction process was first explained by Austrian economist, Gottfried Haberler, in an early edition of his "Prosperity and Depression," (1937; 1939). I should add, however, that Haberler (like Pigou) did not advocate falling prices as a means of re-generating spending in an economy.

And for a contemporary piece that brings out the self-correcting nature of the market in the context of Say's Law for partly understanding the end to the 1920-1921 downturn, see, J. Laurence Laughlin, 'The Industrial Outlook,' in "The Journal of Political Economy" (April 1926) pp. 209-218.

(Laughlin founded the economics department at the University of Chicago in 1892, and was a classical liberal and strong advocate of laissez-faire on a wide variety of public policy issues. He was a critic of the Irving Fisher version of the quantity theory of money, and a supporter of the gold standard.)

Richard Ebeling

I feel your frustration. And it must get worse when the Stephans of the world say that stimulus spending wasn't enough. How much would have been enough? As a non-economist, I am thoroughly confused by the Keynesians and their apologists. On the one hand they say, "Thank goodness we spent what we did or the recession would be worse," and on the other they say, "We didn't spend enough." And on top of this, Keynesian economist Christine Romer predicted unemployment will hit 9%(?)without stimulus and it actually does with stimulus. If we free-market types are skeptical of Keynesianism's scientific pretensions, is it because we live in caves or because the evidence gives us reason to be skeptical? If Krugman argues in a counter-intuitive way that 9-11 was good for the economy, is it also unreasonable to be skeptical of that type of argument too? I understand how waging war generates spending but I don't understand how diverting resources for waging war can be considered as "investment." How can we not be poorer?

The issue is complicated.

If the government did exactly what some interventionist advocated and it didn't work, then the interventionist would be wrong. If the interventionist can arbitrarily vary the details of the explanation to avoid the contradiction, then they just have a bad explanation, since good explanations are hard to vary when confronted with falsifying facts.

The problem is the government rarely does just what anyone in particular advocates. Most people just cross their fingers and hope approximation will work anyway. I advocated QE2 because I figured it would be better than nothing, but I didn't have high hopes. It wasn't the policy I would have preferred in the abstract; it was just the policy I preferred out of those with a realistic chance of happening.

I still think nominal expenditures are too low and that expansionary monetary policy is desirable at this time. I don't think that makes me dogmatic.

Steven. No problem. Be my guest.

As a non-economist, I am confused about the debate between Misesian and neo-Austrians (?) as it pertains to monetary equilibrium and, especially as it pertains to the 1920-21 episode. Perhaps I will forever be over my head, but I find it fascinating, which I why I am willing to go where angels fear to tread.

Richard Ebeling, in his characteristic lucid style, raises what I think is a good point: Why isn't the 1920-21 episode a problem for the neo-Austrians? I learned of the "who goes first problem" by reading Horwitz' book on Austrian macro. I found it convincing that deflation poses a problem for the economy due to prices/wages being sticky downward. But I am nonetheless puzzled and intellectually challenged by the point Mr. Ebeling raises. In a recent post, Bryan Caplan talks about sticky wages as a market failure under a central bank and I agree. But in response, a commentator proffered that sticky wages might be the result of a populace that has been conditioned to inflation. Is it possible that wages adjusted so quickly in 1920-21 because that generation (and previous) was not accustomed to (perpetual) inflation? And, if so, maybe the "who goes first problem" is not as thorny as some think? Is there any plausibility here? Again, thoughts from a non-economist. Be gentle folks!


The socialists argue that the stimuli weren't large enough. But there is no evidence that stimuli have a threshold, like neurons, for which the stimuli must be strong enough or the economy doesn't fire.

And they forget the problem of diminishing marginal returns. In most things economic you get the biggest bang for your buck with the first dollars spent. After that, the returns diminish rapidly.

Finally they'll say it would have been worse without the stimuli. But how do they know? They invented the horror story that the world was ending.

They remind of the guy blowing his horn in central park to keep the elephants away. When a police tells him to stop blowing his horn because there are no elephants in central park, the guy takes credit for the absence of elephants.

If the entire nation starved to death in spite of multiple stimuli except one lone socialist he would say "It would have been worse."

You can lead a socialist to the truth but you can't make him drink.

McKinney -
I can't think of a single socialist that has been involved in any of these debates in any major way. Who do you have in mind when you say "the socialists"?

Some guys made millions off TARP by buying junk bonds and selling them to the government. Others made millions during the real estate bubble and still others have made millions from the Fed fueled stock market.

Someone said of the Israeli/Palestinian problem that the Israelis will always win but the Palestinians will always be there to make sure they don't enjoy it.

Don't be bitter about the fact that socialists always win policy debates. We will be there to make sure they don't enjoy it.

Use your knowledge of the business cycle to make money and help your friends make money. Money is a good salve for being in the minority.

I must agree with fan of rand that Fed policy has contributed to prosperity -- but not here. Cheap dollars have been invested overseas. It is a channel the Fed just will not acknowledge.

To clarify and extend Richard's analysis, the Austrian variant of the cash-balance approach emphasizes relative price adjustments. The totality of those adjustments may result in a decline in average prices. But it is the relative price adjustments that are key.

The difficulty in any counter-cyclical monetary policy to prevent deflation is that it may forestall the needed adjustments in relative prices. Those Austrians advocating stablizing MV are aware of the problem, but feel the harm done is less than the harm from overall deflation.

"And they forget the problem of diminishing marginal returns."

Sounds like marginal efficiency of capital to me--something that doesn't hold perfectly when heterogeneous capital goods are taken into account...

"In all prior donwturns (including the severe contraction of 1920-21), the response was to let markets adjust. And they did, as convetional economic theory predicts."

In the Depression of the 1890s real gdp did not return to its 1892 level until 1897.

Stanley Lebergott estimated that unemployment was over 10 percent from 1893 to 1898.


You picked what is arguably the most serious 19th-century depression. The question is what is the overall record of (1) intervention vs. (2) non-intervention. Selgin, Lastrapes and White re-examine that in "Has the Fed Been a Failure?"

Recent historiography revises downward the frequency and depth of 19th century crises and downturns. They review that literature in the paper.

When GDP is back at its 1984-2007 trend, but real GDP and employment remain 5% and 10% below trend, then I will admit defeat.

I am not sure how much base money can increase without GDP increasing so that I will accept the money doesn't matter business. I guess there is something almost ideological about by refusal to accept that the demand to hold money just always passively adjusts with the quantity of money.

As for Ebeling's point--who is really claiming that a sufficiently large decrease in prices and wages won't do the trick?

The question is whether there is any real advantage to such an approach when it is so easy to increase the nominal quantity of money.

(On the other hand, I think the lower price level approach is less effective with a fiat money than with a gold standard.)

On the 1920-21 episode, see:

Daniel, I consider "liberals" and "progressives" to be socialists. I can't bring myself to call them what they want to be called. It's dishonest.

Brad, diminishing marginal returns applies to a lot of things, revenue, utility, taxes, money supply and state stimuli. But state stimuli have nothing to do with capital.

Bill, so the only way you'll give up on monetarism is if inflation reaches 10%, the gap between ngdp and rgdp?

One of my problems with Sumner, among many, is that he defines tight money by ngdp growth: if ngdp doesn't grow, then by definition money is too tight. He leaves no room for the idea that monetary policy might be loose but not be able to raise the money supply significantly.

He follows a hydraulic concept of economics. People don't play a role. If the Fed really wanted money to grow it would and what people wanted would not matter.

Bill: "The question is whether there is any real advantage to such an approach when it is so easy to increase the nominal quantity of money."

If it's so easy, why hasn't the Fed done it? The money supply has been growing at a pretty good clip. Why hasn't ngdp followed?

Bill Woolsey,

If Reinhart and Rogoff are correct, don't expect a return to computed trend growth anytime soon. Trend growth falls after major financial crises.

Money growth doesn't address the deleveraging taking place. Creating money doesn't create savings. Savings are what rational economic agents want more of.

I agree the real balance effect is stronger under a gold standard.

I understand McKinney's frustration, but caution against his loose use of "socialist." Socialism has a reasonably precise meaning: a belief in state-ownership of the "Commanding Heights" of the economy.

Like Daniel, I don't see real socialists as playing a major role in the current discussion. Nor do I think President Obama is a socialist. He has no sophisticated ideology. His administration has a strong bent for corporatism, however, which is something different.

Just a few data points to support my earlier comment:

As you will see S&P 600 Net Income has not suffered at all under QE1 & QE2, yet unemployment has risen significantly.

Couple more data points on bond issuance under QE1 and QE2:

Corporate bond issuance has also been strong

So the unless production and investment is being expatriated, how else do you reconcile the growth in corporate profits and debt?

Long and short of it, the BRICs and less developed economies are enjoying the benefits of the QE carry trade, where, effectively, money is lent, in this case to American companies, at low interest rates and invested in other countries where interest rates are higher.

Under QE1 and QE2 corporate CEOs enjoy more generous salaries, bonuses, restricted stock units (RSUs) and stock options based on improving returns on invested capital (ROIC). Wall Street firms continue to enjoy not only investment banking and underwriting fees, but also trading fees from transactions with the Federal Reserve to effect the QE1 and QE2 programs.

And US workers draw the short straw because their jobs are outsourced or replaced by cheap labor.


Correction to my last post:

That should have read S&P 500. (Fat fingered 600)

I agree with Lee Kelly. I made a post about a similar topic earlier today. I don't know on any theoretical point I disagree with Professor Horwitz on, but my reading of recent events is almost the opposite. I just don't find the alternative reasons why we're still in the crisis to be at all convincing when you look at different countries. If regime uncertainty is the problem, why is it the problem in the US and Britain, but not Sweden or Australia? The only thing that the countries that are doing okay have in common is that they stabilized expenditures. The Fed didn't, and that is true no matter how many dollars it printed.

Not that it matters, but the use of "socialist" to describe many garden-variety left-liberals is just fine in the casual vernacular, if not a good technical definition. Indeed the policies of old Labour in Britain were widely viewed as "socialist", and inasmuch as Obama fits into this type of mold, I think the label seems to fit fine. Curiously it also seems to fit many Republicans too.

On a more important note, I am glad to finally see some serious discussion with the quasi-monetarist types such as Woolsey, Sumner, and Beckworth.

The idea that the Fed might engage in aggressive open market operations in the face of a sovereign debt downgrade, even if to place nominal money expenditures back on a pre-2008 growth path, seems absolutely nutty IMHO, and suspiciously like an intentional juggling trick. I cannot see why this will not be viewed by the bondmarket as a blatant juggling trick considering the downgrade.

If the Fed chooses to go to another round of QE, then it should do so secretly so as to prevent a full knowledge of inflationary policy by a large segment of the public. Such a broad realization could spell disaster and spark the late stages of a crackup boom.

Many need to go back and re-read Mises' contemporary accounts of 1923.

"It is a channel the Fed just will not acknowledge"

Yes. Another reason is that much of the new base money has been sucked up by foreign central banks to sustain crawling pegs.

Interestingly Mises tells the story that in Germany, the inflation of 1923 was initially delayed by the flow of Marks outside Germany.

Jerry: "Socialism has a reasonably precise meaning: a belief in state-ownership of the "Commanding Heights" of the economy."

What about European socialists? They decided before WWII that a small free market was necessary for them so survive. Seems to me that European socialists don't want to own things as much as they want to control them, like American "liberals" and "progressives."


"Recent historiography revises downward the frequency and depth of 19th century crises and downturns."

Recent estimates do significantly revise the severity of downturns but not so much the length.

Recent estimates of industrial production by Joseph Davis still show that production took about 5 years to recover.

Not all historical recessions were as short as 1921.

In addition, it is not clear to me that we can divide U.S. history into a period with intervention and a period without. Monetary policy, tariffs, and an income tax (overturned by the Supreme Court) were central economic issues in the 1890s.

Ksralla, I refer to the left leaning Republicans as socialist-lite.

The strange thing to me is how the left in the US consider "socialist" to be an insult. You won't find many people in Europe who think it is an insult. And yet what American "liberals" and "progressives" believe is exactly what European socialists believe.

Some think they're not socialist because they don't advocate state ownership, but merely state control of businesses. What is ownership if not control?

The National Socialists fooled a lot of gullible people by leaving the paper title to companies in the hands of the owners while taking control of every aspect of the companies. That way they could claim that the state didn't "own" the business.

In the US, the National Register increases on average 70,000 pages per year. Regulation costs businesses over $1 trillion.

Corporatism is nothing but a variety of Marxism. Remember Marx left vague the outlines of a socialist economy. The USSR tried one variation; France, Germany and Brittain tried another; Italy tried a third implementation. But all had the same goal: state control of business.

fanofrand: "American companies borrowed money (at low interest rates)and with it purchased equipment manufactured in the BRICs or less developed economies or built factories in those countries..."

Wouldn't that keep the US money supply from growing? Instead, it has grown at a pretty good clip.

fanofrand: "...where labor costs are a fraction of US labor costs, health care and other employee benefits barely exist and labor standards look like they were drafted from a Dickens novel."

Actually, if you use total labor costs (adjusted for productivity) instead of wages, US workers are among the cheapest in the world. Taxes and regulations are the main drivers of the US companies' failure to compete.

Bob: ""What do you mean, TARP failed?" They claim the gov't made money on it, it resuscitated asset markets, etc."

And that's what they say. I remind them that the original goal wasn't for the feds to make a profit. The goal was to decrease unemployment.

Socialists like to think they can pin us into a corner by claming that we think government action has not benefits whatsoever. I remind them that is a straw man. We do see benefits from state action; we just compare those benefits to the actual goals and to the costs.

We try to do cost/benefit analysis; socialists try to get by with benefit/benefit analysis.

Socialists hate big business with an irrational passion. Yet whom did TARP benefit? It boosted the largest banks while screwing over the regional banks. They never consider the costs.

One last post and I'll quit hogging.

I visited one of the top socialist web sites recently and they addressed the issue of state ownership of businesses. They said the atmosphere in the US was unfavorable toward state ownership, so while they would still beieve in it they wouldn't advocate it. Instead they would promote state control of the market, since many Americans like that.

And they understood that state control through regulation would eventually lead to state ownership.

The TARP point is odd because a lot of the people that are regularly criticized on this blog didn't like TARP, said in advance it would create zombie banks, and supported the Swedish solution.

"It made a profit" is something I only really hear from Democratic Congressmen. I guess that sweetens the pot marginally, but I don't really care about that and I don't think most people do - despite Bob's claim.

I'm no expert on financial markets - I would guess TARP ended up being a better solution than doing nothing at all, but it was considerably inferior to the sorts of resolution options to prevent zombie banks that LOTS of people were advocating.

Brad, you should check out Christine Romer's revisions of the late 19th century business cycles. Very interesting.

I have some thoughts here on the other three assertions by Steve:

Ryan Murphy,

"If regime uncertainty is the problem, why is it the problem in the US and Britain, but not Sweden or Australia?"

Have the central banks of Sweden and Australia rescinded on their declared policies? The Bank of England effectively has, inflation targeting is a piece of surrealist theatre.


"He [Sumner] leaves no room for the idea that monetary policy might be loose but not be able to raise the money supply significantly."

In some ways I can't believe that people on this blog are making this argument. This is basically the Keynesian argument for why monetary policy may be ineffective. It's an argument for fiscal policy!

It's also mostly rubbish. I have described why in mind-numbing detail on the Cobden Centre site recently...

There is no reason why interest on reserves can't be removed. There is no reason why bankrupt banks can't be liquidated, and new banks allowed to take their place. If Walmart were permitted to become a bank I bet they wouldn't hold as much in excess reserves. Monetary policy can be effective if the government really want to make it so.

Lastly, in extremis there is no reason why the central bank can't charge interest on excess reserves.

I agree with Jerry O'Driscoll however that there is a great deal of balance-sheet repair going on here that can't be assisted by monetary policy.

Daniel Kuehn,

If anything this debate shows that nobody in economics uses direct empiricism, even if they claim that they do. Nothing like a resolution to these debates will come until there is good evidence on the microeconomic facts that each theory relies on.

McKinney, I don't understand either of your two comments about my post. QEs have artificially reduced the cost of money. How does companies issuing new debt shrink the money supply?

And as for your comment that American workers are the cheapest, taxes notwithstanding, please provide actual data supporting that.

And, also, please clarify what you mean by "regulations". Does that mean labor laws as opposed to the oppressive working conditions that manufacturer's in less developed countries can operate under with impunity?)

As I noted in my original post, a little less subtly than i will do now, the greatest US manufactured product in the last 10 yrs is the fantasy that manufacturing has moved off our shores due to taxes.

It will be impossible for America to recapture its luster without rebuilding a manufacturing base.

(BTW, i don't work in manufacturing sector so this is not a self-interested argument. In fact, I work in the very global telecommunications industry that has greatly facilitated the ability for companies to move their manufacturing, procurement, services, etc. activities to cheaper markets.)

current: "Monetary policy can be effective if the government really want to make it so."

Then why isn't it? And what about the long lags?

current: "this debate shows that nobody in economics uses direct empiricism..."

The direct empirical evidence says that the lags between policy and effect are long and variable. Friedman and Greenspan didn't make that up out of thin air.

The fact that monetary policy doesn't work is not an argument for fiscal policy unless you think fiscal policy can work.

fan: "How does companies issuing new debt shrink the money supply?"

Not the issuing of new debt. Obviously that increases the domestic money supply, but only if the new money stay in the country. If it leaves the country, then the domestic money supply shrinks.

fan: "And as for your comment that American workers are the cheapest, taxes notwithstanding, please provide actual data supporting that. "

It's all on the BLS web site.

fan: "Does that mean labor laws as opposed to the oppressive working conditions that manufacturer's in less developed countries can operate under with impunity?"


fan: "he greatest US manufactured product in the last 10 yrs is the fantasy that manufacturing has moved off our shores due to taxes."

US corporate taxes are the highest in the industrial world. You don't think that hurts competitiveness?

The smart socialist countries, like Sweden and Denmark, have reduced corporate taxes while increasing personal taxes in order to make their businesses more competitive internationally.

My point has been simple, right or wrong, QEs have had their desired effect of keeping interest rates lower than what they would have been without the Fed's intervention. However, the benefit of those Fed activities has been enjoyed by workers and economies outside the US.

The QEs have effected the carry trade of all carry trades, resulting in cheap borrowing costs, greater foreign investment in the BRICs and LDCs and accelerating the US's lack of competitiveness.

Daniel: "The solution to getting an unbiased estimate of the impact of stimulus is to identify variation in policy that is exogenous to what's going on in the macroeconomy and then look at the association between that portion of the variation in policy and the behavior of the macroeconomy." (from your blog).

No that is not a solution. The solution does not like in empirical data at all. Anyone who has done Design of Experiments knows why the predictor variables have to be orthogonal: you can assign cause only to orthogonal variables. Any significant correlation between variables causes confounded effects, meaning you can't determine which one is causing the effect.

Historical data is highly, highly correlated. So statistically it's going to be impossible to determine cause/effect. Those that seem to are usually guilty of leaving out correlated variables.

You can't do controlled experiments in macro economics, so statistical analysis for cause/effect is worthless.

So how can we say that TARP and QEx failed?

1) They failed to do what the proponents claimed they would do - reduce unemployment.

2) We have historical data for periods in which neither monetary policy nor fiscal policy were used, all pre-1929. We can compare those periods with post-1929 experiments with fiscal/monetary policy. If fiscal/monetary policy worked better than no fiscal/monetary policy, the results should be statistically different using something as simple and ANOVA. They aren't.


If you have some time check out this paper Anthony Evans and I have written over on SSRN:

Part of the problem with Friedman's version of the quantity theory MV=PY is that it concentrates on output, not total transactions. I think that confuses the picture.

I think that if we were to use MV=PT then the situation would be clearer. That is, when M is increased that bids up the constituents of T. I think the lags occur mostly because of the market in assets. Specifically, if asset prices are bid up then it takes some time before that produces a rise in GDP output prices. In my opinion monetary equilibrium is when we have stable PT or a stable PT trend, not stable PY.

You're right though that even by this sort of criteria lags may exist. I believe the answer to this lies in anchoring expectations. Agents must be convinced that monetary equilibrium will occur. If they are then shifts in both demand and supply of money will be smaller.

McKinney, the BLS site has tons of data. Can you please provide an actual link to the data that shows that US workers are cheaper cause I don't see that?

On the issue of oppressive work conditions abroad, which you seem to think is ok, you seem to be intimating that US corporations should be permitted to operate under the same model. Am I understanding you correctly?

Help me understand where your line is when it comes to acceptable working conditions and the corresponding "regulations" because that is really key to your argument. For example, are blood diamonds ok? How bout mandatory 60 hr work weeks? How bout the right to fire women who are pregnant and forcing woman to take pregnancy tests? Or encouraging or incentivizing
with money co-workers to inform managers of pregnant co-workers so they can be summarily dismissed? My problem is that at 10k feet and 7k miles away we don't have to think too carefully about oppressive work conditions in these countries. All we know is that a computer component can be produced at a fraction of the cost as it can be here in the US. But when it's your wife or daughter, for example, that is subjected to those conditions, my guess is your tune would rapidly change.

I agree that taxes play a role in investment decisions ONLY insofar as the NPV of the project changes based on that variable. My point is that the tax variable alone hardly offsets the cheaper labor costs in these countries.


By "intervention," in context, I meant counter-cyclical policy. Before the Fed, there were no serious interventions. Downturns ran their course.

In 1920-21, Harding ran promising to let markets adjust through deflation. Hoover was the big break.

fan of rand would rather people starve to death than that they work under conditions of which he does not approve. By all means, proudly proclaim a victory over child labor while ignoring the spike in child prostitution that follows. We are so wealthy because the West went through situations we now condemn -- and in so condemning, prevent others from reaching our position. But those are just brown people, so the heck with them, right? So long as we get manufacturing back in this country, right? And never mind that we have a lot of manufacturing in this country, or that in many ways we are moving into a postindustrial economy (which will be wealthier, by the way -- if you believe Richard Florida and all his evidence).

All of the things our government did to "help" the economy only helped large companies -- at the expense of the startups that actually do the hiring of new people. Large companies do in fact tend to outsource (small businesses cannot yet do so) and, more, find more efficient ways of doing things, so that fewer people can make more. I think this latter is what we have been seeing over the past few years.

I don't like the term "loose" or "tight" money, but I think it is obvious that there can be "loose" money without the quantity of money rising. It happens with the demand to hold money falls. There is an excess supply of money.

Current, perhaps the amount of money necessary for transactions given higher asset prices is important, but generally, I focus on the difference between the asset yields and the yield on money. Higher asset prices, lower yields, lower opportunity cost of holding money, demand for money rises to clear up the excess supply. (Higher asset prices, the ordinary churn in asset markets requires more money at higher prices, the demand for money rises, the excess supply of money is cleared up. OK...)

Fan of Rand is falling into the most common fallacy in macroeconomics. We have this long term trend of economic development in poor countries, and this occurs with employment and output growing just fine in the U.S. Then, when we have a recession, suddenly, that is what is causing our problems. Maybe, but not likely.

The best solution of monetary policy lags is a clear committment for a growth path for what you are trying to target. GDP for quasi-monetarists.

If people believe you will keep on trying until you get there, then their expectations of higher GDP in the future will raise in now. (And, simiarly, if they believe you will contract to get it back down, they there expectations that it will go back down, or grow more slowly in the future, will slow it.

Looking at how GDP responed to the M2 measure of the quantity of money, when the Fed was targeting interest rates someone contrained by gold, though with huge gold reserves... what does that really tell us about what happens when there is nothing that requires GDP adjustments in equilibrium. The quantity of money isn't exogenous in such periods. (And, of course, it isn't exogenous with GDP targeting.)

What happens if the quantity of money rises, and nobody knows what will happen to the quantity of money in the future, or GDP? How long will it take for GDP to fully adjust? Or even start adjusting?

As for history, if the demand for gold rises faster than the supply, ceteris paribus, GDP needs to fall. But, in the real world, the Fed would have to notice this and raise target interest rates. When will that happen? And, of course, then can release gold reserves for a while. How persistent is the change in fundamentals of the gold market? But, the Fed gets worried with its gold reserves are low. Can they convince other central banks to support their policy. You look at history from that world and you see that GDP only follows money with variable lags?



I favor GDP targeting that at most would result in 7% inflation over the next year and certainly would move to minimal inflation afterward. Your view is that foreign bond investors will think it is inflationary because they believe that the long run price level is proportional to base money?

As for these bubbles, the problem is taking short term interest rates and projecting them into the future--a mistake. Don't do it.

Or worse, assuming past price changes of an asset can be projected into the future. Don't make that error.

But I think stabilizing short term interest rates because foolish investors might project them into the future is a mistake. Keeping asset prices from rising or falling because fools might think that means they will rise or fall more is also a mistake.

Slow, steady growth in money expenditures on output is the least bad macroeconomic environment for microeconomic adjustments.

Deleveraging is just a twist on the paradox of thrift.

Deleveraging is saving. The notion that nominal expenditure much less real output and employment must be lower if people save more is false.

All that is necessary is that real interest rates be at the level that coordinate saving and investment. Nominal expenditure can continue to grow and real expenditure equal the productive capacity of the economy.

The fallacy is to assume that those who receive debt payments simply accumulate money balances. They might, but increasing the quantity of money to accomodate any such desire is how money address the problem.

At the same time, to the degree this increase in saving requires a lower real interest rate, real market interest rates must decrease. If the zero nominal bound is hit on some financial assets, what that means is people hold money rather than lend. An increase in the quantity of money addresses that.

If open market operations are limited to the sort of short and safe assets that have hit the zero nominal bound, then that sort of monetary policy won't work.

Either you have to break the zero nominal bound with negative nominal interest rates(which requires privatization of hand to hand currency as a practical matter,) or price inflation must be targeted to lower real interest rates despite zero nominal rates, or the money issuer (like the Fed) has to buy long and risky assets rather than short and safe ones.)

But again, "deleveraging" is nothing more than the paradox of thrift. It only implies lower money expenditures on output if we assume in directly or indirecly creates an excess demand for money, and monetary policy can address it.

I don't think having price and wage deflation and then reflation is a sensible approach to "deleveraging," and reducing output and employment is no "solution" but just a tragic waste.

US direct investment abroad peaked in 2007 and has fallen since then. It is about $220 billion which is ~1.56% of US GDP. Most of it is invested in other developed countries not developing countries. See:

I must declare an interest here.... I work for the Irish branch of a US semiconductor company, I'm an engineer. I could tell you a lot about pricing of computer components.

I agree in principle that export of capital from a country disadvantages unskilled and more lower skilled workers. I'm not convinced that it's a very important at present though.

One correction:

Real GDP is 12% below the trend of the Great Moderation, it is 7% below the CBO estimate of potential. Civilian employment is 10% below the trend of the Great Moderation.

The price level is 2% below the trend of the Great Moderation.

"I favor GDP targeting that at most would result in 7% inflation over the next year and certainly would move to minimal inflation afterward"

OK Bill. So it's now a very modest 12% inflation?

Bill, I am not foolish enought to become embroiled in an argument on macroeconomic theory with you. You will win. I read your blog comments regularly (here and other places), and must confess that I struggle to follow your thinking quite often.

However, the fact that you have the balls to make these quantitative inflation predictions reveals too much about your monetary worldview. The fact is that you cannot know how the market might respond to an injection of the magnitude required to return NGPD to it previous trend line. You cannot know the effect that this has on the confidence of individuals that their money is still good. So in the face of a downgrade in U.S. debt, I stand by my contention that on its face, your suggestion is nutty.

I have no anger for your view, only curiosity of how you have become so darn sure you are correct.

I've taken my valium and am a bit calmer now.

NGDP targeting is a great idea. Yet it is the smug confidence that your bunch gives off, that communicates that any human being possesses enough knowledge to skillfully predict the outcome of the techniques required to achieve a certain NGDP trend that rightfully deserves some mild indignation.

I am familiar with your CV, and with the academic pedigree that you are a part of, your confidence surprises me.

I normally like to find a point of agreement in someone's comments and build on that. I'm hard pressed to do that with Bill Woolsey's response.

I will note the irony that Austrians were recently accused of believing that savings automatically translates into investment (untrue). Now Woolsey accuses us of ignoring how adroit monetary policy can bring up about just the right interest rate so that savings and investment will be coordinated.

Deleveraging is a real adjustment, which can take years. Reinhart and Rogoff have estimated as long as a decade. Changes in a nominal variable (money) cannot effect a real adjustment. That is done through changes in relative prices, including a whole array of intertemporal prices. The idea that changing the quanitit of money is a substitute for all these relative price changes is simply a fantastic notion.

One aspect of the needed adjustment is that there is tremendous excess supply of housing. It is not just that "too many" houses have been built, but they have been constructed in the wrong locations. The counterpart of that malinvestment in housing is that there has been too little investment in real capital elswhere. Does Woolsey think you can conjure up real capital by printing money?

Rogoff wants to inflate away the excess debt. If Woolsey wants to join the inflationist camp, I wish he'd just say so.

I think I said, "at most, 7% inflation."

I hope increasing GDP 10% over the next year would result in about 10% growth in real GDP. Potential output continues to grow, and the output gap can close.

But if there is no output gap to close, then the 3% growth in potential and 10% growth in GDP gives you 7% inflation.

The reality is likely to be somewhere in between.

Anyway, once on a 3% growth path of GDP, if potential grows at a bit less than 3%, as predicted for the next few years, then inflation will be very low. Well under 1%, close to .5%.

(What I don't pretend to know is what the appropriate levels of the quantity of money are to keep GDP on target.)

K Sralla said:

"Not that it matters, but the use of "socialist" to describe many garden-variety left-liberals is just fine in the casual vernacular, if not a good technical definition. Indeed the policies of old Labour in Britain were widely viewed as "socialist", and inasmuch as Obama fits into this type of mold, I think the label seems to fit fine. Curiously it also seems to fit many Republicans too."

This is a terrible analogy. Old Labour was, by its own description, a socialist party. It carried out wholesale nationalization of many industries, and transparently catered to the wishes of unions. The change from "Old" to "New" Labour happened when they dropped support for wholesale nationalization from their party platform.

Obama, by contrast, conspicuously did not nationalize even the health insurance industry, he did not nationalize a single bank when some were urging him to, and did not aggressively push the reforms that unions favored. The number of government employees today is lower than it was in 2008. "Socialist" is a wildly inaccurate characterization of such policy.

A 7% inflation rate confiscates one-half of the value of nominal assets in about 10 years. Sounds like communism on the installment plan to me.

Jerry, I'm stealing that comment as a quoted Facebook post.

Jump back up the thread a bit to Richard's comment about MV, don't you think we should distinguish between a theoretical ideal in which a free banking system will tend to stabilize nominal spending (White, Selgin, Hayek) and belief that a central bank can effectively stabilize nominal spending (Sumner) and the notion that, if you must have one of these rotten stinking central bank,s aiming at nominal income stability is better than targeting interest rates or 2% inflation or whatever the hell it is that they claim to be doing now?

I'm out of my depth on this whole issue, but there does seem to be some conflation of theoretical ideals with fed policy recommendations that makes understanding these MV-related discussions pretty hard to follow.

Ann Coulter, "Slander", Chapter NINE​nt/pub?id=1GD-bpXvucQbIuPDP44u​a1nYFZqOiCDghozcICaFLBuo

Jerry, John,

It's not 7% inflation over the next ten years. It's (maybe) 7% inflation over the next years, and then no inflation, on average, from there on out.

Correction: that's "It's (maybe) 7% inflation over the next year" rather than "years."

The point is that Bill Woolsey is advocating a one off spurt of inflation rather than high inflation for the next ten years. Once NGDP has returned to Woolsey's preferred trend, average inflation will fall to zero. In other words, in the long run, we'll have slowly rising NGDP and a stable price level.

McKinney -
re: "Anyone who has done Design of Experiments knows why the predictor variables have to be orthogonal: you can assign cause only to orthogonal variables."

Ummm... right. What do you think IV approaches (however imperfectly implemented - of course I'll admit to that) are doing?

Daniel, but IV don't do the trick. IV can only isolate one or two factors, and then not completely make them orthogonal, when there are many correlated factors, in addition to the fact that there are many important, correlated variables left out of most models to keep them simple.

Daniel, if the process is a simple as you think it is, why are there so many contradictory models of the latest crisis? There are as many models "proving" that TARP and QEx failed as there are showing it worked. Do these modelers not understand IV?

I have asked proponents of ngdp targeting several times how they get around the problem of long lags and have yet to get an answer.

Sumner says they don't exist because the market reacts instantly to policy changes. Clearly the financial markets react instantly, but if the goal is to reduce employment and boost ngdp, there are still long lags between financial market reaction and changes in ngdp.

Sumner has removed the long lags, he has only succeeded in clouding the issue and pushing back the lags by one step.

Daniel, If I remember my DOE, adding IV doesnt reduce correlation, it reduces error. The only fix for correlation that I'm aware of is factor analysis, structural equation modeling, or partial least squares.

Lee explained it correctly.

I don't favor 7% inflation for 10 years. But I don't even favor 7% inflation for any years.

But if we did have 7% inflation for 10 years, I would expect the nominal interest rates on nominal assets to all rise to something greater than 7% and so hardly any nominal assets would lose 1/2 their value in 10 years.

And even if we did have transfers from creditors to debtors because of inflation, that wouldn't be socialism.

And to the degree this inflates away part of the government's debt, it is a tranfer from the government's creditors to taxpayers.

Even unanticipated inflation doesn't transfer from the private sector to the government.

And, of course, the deflationist approach is about engineering a transfer from taxpayers to government bondholders. Not socialism, of course, but it is something you all should keep in mind.

I favor 10% GDP growth over one year and then 3% GDP growth after that. It is catching up to a growth path over the next year and then staying on that growth path.

If I am right, then the inflation rate will slow--moving towards .5% and real output and employment will increase rapidly, and then begin to grow more slowly. That low inflation follows from estimates of below trend growth in potential output over the next decade. (Closer to 2.5% rather than 3%.)

If we can get some good supply side policies, then the low inflation would move towards zero, and maybe even to a slight deflation.

If I am completely wrong and output is currently at potential and the level of prices and wages are at equilibrium already, then the result would be 7% inflation this one year, but then it goes to the .5% as before.

If we had GDP targeting at a 5% growth path (which I never liked,) and it is really the case that potential GDP fell 12% below its previous growth path, then we will have had higher inflation over the last 4 years. Maybe 4% in 2007, then in 8% in 2008 and 6% 2009. With a stablized MV, sharp decreases in productivity lead to sharp increases in prices.

If GDP were growing 3% all along, then the result would have been the same, but just lower inflation. No inflation in 2006, 2% in 2007, 6% in 2008, and 4% in 2009. And then back to .5%. Or something along those lines.

I don't think potential GDP dropped like that at all. And so, I think if GDP had stayed on the 5% trend, maybe we would have had some 3% or 4% inflation and real GDP would have continued to grow, just 2% or 1%.

With my prefered option, it would have been lower inflation--1% or 2% inflation, when real GDP grew only 2% or 1%.

Oh, and by the way, if real GDP starts on a downward spiral of shrinking 4% each year, so that 3% GDP growth results in 7% inflation for a decade, then I think that nominal assets dropping by half in real value every 10 years is the least bad option. Of couse, nominal interest rates might rise, but.. who knows in such a horrible scenario.

P.S. Please keep in mind--the monetary equilibrium approach isn't about always increasing the quantity of money. It is about decreasing it when appropriate as well. When the demand for money falls, the fixed quantity of money approaches generate a higher price level and at least transitional inflation. When the demand for gold falls, what happens to nominal wealth then?

With all this failure asset prices as measured by the S&P 500 nearly doubled as these policies took effect. I'll take more failure like that!

Jonathan, agreed! There's some significant coin to be made on the Fed's moves.


> Higher asset prices, lower yields, lower opportunity cost of
> holding money, demand for money rises to clear up the excess
> supply.

I don't think it can be rectified as easily as that. But that's a question for another time.

I agree with you to some extent that debt deleveraging is like the paradox of thrift. Jerry and I were just talking about the savings-investment market in the previous post, as he mentioned...

> The notion that nominal expenditure much less real output and
> employment must be lower if people save more is false.
> All that is necessary is that real interest rates be at the
> level that coordinate saving and investment. Nominal
> expenditure can continue to grow and real expenditure equal the
> productive capacity of the economy.

I agree in the ideal case, but I can't agree that all cases are ideal.

You wrote:
> The fallacy is to assume that those who receive debt payments
> simply accumulate money balances. They might, but increasing
> the quantity of money to accomodate any such desire is how
> money address the problem.

Three days ago I made pretty much the same argument:
> The incorrect part concerns money. Suppose I hold a bond, and I
> decide that it's too risky so I sell it. That means I then hold
> money instead of a bond. If I keep that money then that means
> my demand for money has increased, in that case we have a
> monetary equilibrium type problem. In that case my bank may
> increase it's lending due to my increased lending to it. If I
> spend the money on investment goods then I have invested and
> there is no problem. If I spend the money on consumption good
> then I've dissaved and consumed, and there is no problem. If I
> spend the money on a different sort of existing investment then
> the person who recieves my money is in exactly the same
> position I was in the start of this paragraph. We can't really
> get to the problem by starting here.

But, this doesn't mean that there isn't a way that this may happen that isn't connected to monetary equilibrium:

> The correct part concerns a shift in the relative demand
> between existing assets and constituents of GDP output. Instead
> of selling a bond suppose that I earn £1000 and I save it in a
> bank account. The loan officers in my bank are too cautious to
> chase new business, so they invest in existing bonds bidding up
> their price. This is what Keynesians call "leakage from the
> output flow". This could also happen directly if I were to
> invest it in existing assets myself. That is, there may be a
> shift in demand between outputs and existing assets in favour
> of existing assets. To the extent that some of those existing
> assets are not reproducible quickly that may produce a decrease
> in investment and therefore in GDP.

One of the ironies about this discussion is that it was you who first got me thinking this way, in a post on your blog on ABCT.

I agree with you that different future price trends have different redistributive effects, and all that is very politically contentious for good reason.

I'm not sure I understand your justification for the NGDP targeting regime you choose. One way to justify a price-level trend type regime is to say that people have become accustomed to a particular trend in nominal prices and plan around them. Another is that they become accustomed to a particular trend in nominal income. As I understand it the way Scott Sumner justifies it is through the real burden of nominal debts. If people expect an income trend and that trend occurs then the real burden of nominal debts doesn't change much.

I don't see how that sort of justification avoids the sort of logic Jerry and I have mentioned. Surely it involves the assumption that the creditors won't behave in the same way as the debtors. That is, if future real rates were high then more real income would go to creditors from debts, and if they were lower then less would travel that way. In this case the only way we can reach a different GDP output is if these two groups have different propensities to spend on GDP output. But, if that the case then why can't there be endogenous changes in tolerance for debt, or desire to save?


The constant harping on deleveraging is vulgar Keynesianism. Deleveraging is increased saving, and increased saving does not require decreased nominal expenditure on output or decreased real output.

I grant that Rogoff appears to agree with the vulgar Keynesian view on the matter. So does Krugman. Any other mainstream Keynesians you want to cite?

Is deleveraging (or any increase in saving supply,) a real change that cannot be avoided by nominal changes? Well, a change in the _real_ interest rate is a real change and that is what coordinates real saving and real investment.

I don't favor interest rate targeting by the monetary authority. You need to think about why you assume that I was claiming that Austrians don't understand that the monetary authority could set interest rates just right to coordinate saving and investment.

I think that most Austrians do understand that an increase in the supply of saving results in lower interst rates and an increased spending on capital goods. Sure, many talk about lower time preference and more round about methods of production, but it amounts to the same thing. There is a shift in the allocation of real expenditure and in the allocation of resources between production for the present and the future.

Now, you can pull out the magic wand of capital heterogenity, and I won't disagree. But that doesn't mean that the complaint about deleveraging is anything more than vulgar Keynesian economics. I will agree--an increase in saving will require a reallocation of resources that can disrupt current production. The allocation of labor, the skills developed by the workers, and the specfic composition of capital are likely to be wrong. But how much will output be disrupted. Not much, I think.

If the quantity of monety adjusts to meet the demand for money, then the supply and demand for credit will generate _real_ market interest rates that will coordinate saving and investment, even if some or all of the saving generated by the dreaded deleveraging. Nominal expenditure can continue on its previous trajectory.

Frankly, your remark that I would believe that money creation can conjure up the capital goods that were not produced because houses were produced is insulting. I can understand that you don't pay attention to what I write, but I have said over and over that misallocation of resources because of the housing boom reduced the productive capacity of the economy. More in the short run, and permanently to some degree.

I would note, (again) that the CBO estimates of potential income suggest that about half of the shortfall of real GDP from trend is due to reduced productive capacity.

Of course, they might be wrong. GDP targeting isn't based on calculating an output gap. If productive capacity is depressed for any reason, including an allocation of labor and capital set up to produce something other than what people want to buy, then the price level rises. As appropriate capital goods are produced, people find new jobs, and improve their skills, the price level goes back down. And yes, all sorts of relative prices are change too.


I thought you explanation of monetary disquilibrium was great.

I thought your analysis of how interest rates might fail to coordinate saving and investment (and your conversation with the postkeynesian) balled up income and substitution effects. But I nearly always suffer from the orthodox failing of focusing solely on the subsitution effects.

My view is about the quantity of money and the demand to hold money, and how that interacts with the flow of production--working, saving, investing, consuming. Asset prices are an afterthought. How do they impact what is important? Obviously, if the goal is to understand asset prices, my perspective is misdirected.

Steve Horwitz,

I think something everybody here has missed is that the baseline of late 2008 was 1931, at least that is what Ben Bernanke thought, a leading scholar of the Great Depression, and I happen to agree with him. If that is the case, and certainly you and everybody else who has commented here so far is free to disagree, then some of the policies that you simply label as "fail" were far from it.

So, TARP (plus the unmentioned stress tests in early 2009 that marked the bottom point of the stock market), indeed prevented a major wave of bank failures as happened in 1931. And, yes, TARP has made money for the govenment, whether or not one approves of that.

QE1 should probably be viewed as an extension of these main items, and the basic fiscal stimulus as well, one third of which was tax cuts, btw.

I am willing to grant that QE2 did not amount to much, for better or worse, and it is arguably the case that the problem with the debt ceiling increase is the "unclean" stuff attached to it that is perceived to be likely to do to the US what the austerity plan in UK has been doing to them, rather than the actual debt ceiling increase, with a failure to have raised the debt ceiling almost certainly having been likely to lead to a much worse situation than we are seeing now, unless that was accompanied by an abolition of the ridiculous debt ceiling as well, which the markets would have probably welcomed.

Do keep in mind, folks, it was in 1931 that the unemployment rate in the US went from 8% to 15% as the greatest financial crash in world history brought down banks all over the world, which in the absence of an FDIC meant the destruction of the wealth of many people whose deposits in those banks simply disappeared. One can go on all one likes about Hoover favoring unions or Smoot-Hawley, but it was this crash that turned what had been an annoying recession into the Great Depression, and the policies carried out in late 2008 and early 2009 probably did prevent us from repeating that unfortunate history.

If we use 1931 as a baseline, then this is a terrible fail. After reaching 15%, unemployment then dropped rather quickly. However, in this recession unemployment shot up to over 10% and has never gotten below 9%. And there is little indication that it is in fact going anywhere. More than that, we are using questionable methods for determining who is unemployed. We only count those who are getting unemployment, which means those who do not qualify or no longer qualify are not counted. The real unemployment rate is likely 15% and stable there. A large number of banks did go under -- but those managed to stay out of the news. After all, most of the big ones were kept safe, and have even had numerous small banks handed over to them (thus exacerbating the "too big to fail" problem).

By all means, call this a success. A three-legged horse crossing the finish line is also a success in the Kentucky Derby by your definition of success.

Sorry, Troy, but the unemployment rate kept on rising until 1933, peaking in the US at 25%. Much worse than what is going on now.

The use of the Great D to bolster one's theory is nothing but evidence that Mises was right: historical evidence is so vast and contradictory that one can find support for any theory in it.

To properly interpret history one needs sound economic theory to make sense of it.

Yes, it peaked in 1933 at 25%, then dropped down in a few years to 8%. Then it rose back up to about 15% because of FDR's policies. Like I said, I don't trust the government numbers on unemployment. This economy also has widespread underemployment and misallocations of human capital on a scale not possible during the Great Depression. Those elements are damaging to the economy long-term. These are the things our policies have, to date, given us. Again, call this a success if you want. It's a dismal failure by practically any rational definition. The only reason this depression does not appear to be as bad as the Great Depression is because the country, as a whole, is far wealthier than it was during the Great Depression. That matters.

> I thought you explanation of monetary disquilibrium was great.

Thank you.

I agree with you that the flow of production is the most important aspect in economics. But, other things must concern us if they have an effect. (I'm not going to go on about ABCT here).

As you know I agree with you about a lot, that's why I'm supporting your view in the other thread. This is nitpicking to some extent.

> I thought your analysis of how interest rates might fail to
> coordinate saving and investment (and your conversation with
> the postkeynesian) balled up income and substitution
> effects. But I nearly always suffer from the orthodox failing
> of focusing solely on the subsitution effects.

It was messy I agree, I'll try to sort it out here.

To begin with, when I read your blog I see a lot of concern over trends in nominal income. I'm not sure I understand why you're so concerned with that. That was why I brought up Scott Sumner's view on debt.

As I see it there are two possible reasons why you may be worried. Firstly, you may think that sticky expectations about income apply over quite a long run, and secondly you may be worried about debt like Sumner.

I'll take the first case first... If it were the case that income trend expectations were very sticky then people now could be judging their income trend using a trend from 2008 and before. If their income constantly falls below that trend then their money demand will continually reflect uncertainty about the future. We then have a fall in V and hence a recession. I can't quite see this. Surely at this late stage more than two years since the recession begun people have changed the way they look at their money income. Surely they now judge it's money value on the basis of the recent past during the recession rather than the period before it? I certainly do.

The second possibility is that you're concerned about debts like Sumner sometimes is. The normal concern here is that if that the real value of debts will increase for borrowers then they must spend less. This is only an issue for output if the spending behaviour of creditors is different to debtors. If it isn't then all we have is a redistribution, which is objectionable in other ways but not the same sort of problem.

If the two different parties have a different demand for money then that's a monetary equilibrium problem as I mentioned before. But, this brings up the problem of the difference between creditors and debtors again. Why would creditors have a higher demand for money? Are they more risk-averse? It's certainly possible. Even if this is the problem haven't you argued before that the Central banks can create any quantity of money? In that case the central banks can raise M to offset V.

Which one of these arguments is your case for an NGDP trend?

Anyway, going back to what I wrote earlier. The issue here is what happens if agents want to deleverage on net. The argument you are making is that the savings-investment market can always permit this and that monetary equilibrium is the only problem. I'm not so sure about that. You say that I'm saying that the savings-investment market is not coordinating, I'm not sure that I am.

The issue here is that those who are advancing money for capital may have various tolerances for risk. In particular tolerance is likely to go down in a recession. The market may not, immediately, be able to provide opportunities for low-risk investment. As a result that fresh supply of savings may cause the prices of existing assets to rise rather than causing investment in new capital equipment.

I'm not sure that this I'm claiming that the market isn't coordinating. Let's say we have a "freely reproducible" good, such as a drill press. In that case we could say that coordination occurs when the same number of new drill presses are produced per year as are bought, when the market for new drill presses clears (even this isn't case isn't clear though, what if the market for existing drill presses doesn't). We don't say the same sort of thing about, say, land though. There is virtually no production of new land (taking land in the physical sense). In that case coordination can only mean that the market for existing land clears.

If we take any particular timescale and any particular market the situation is somewhere inbetween both cases. Nothing is perfectly "freely reproducible" or perfectly non-reproducible (Hayek has a good spiel on this in "Pure Theory of Capital"). With investment the issue isn't so much producing physical capital goods, as producing an overall functioning business (or department in a larger business). I think this can always be done in time, but not immediately.

Recently, I've become a sort of "Gloomy Austrian" (or maybe some would say "Gloomy-Austro-Keynesian"). I think that monetary equilibrium is the best we can do to stabilise economies. But, I don't think that it can prevent recessions occurring or continuing in all cases.


If the demand for investment is prefectly inelastic with regard to the interest rate, so that an increase in the supply of saving doesn't impact the production of new capital goods at all, then the result is that the interest rate falls enough so that the quantity of saving supplied falls back to its initial level.

The interest rate is still coordinating saving and investment.

If we imagine the demand for investment being perfectly interest elastic, then an increase in the supply of saving would be transformed into investment, real dollar for real dollar. But that is a special case.

I go with the "usual" case where saving and investment are neither perfectly elastic or inelastic. And so, the amount saved and invest increase, but less than the amount that would be saved at a constant real interest rate.

There is no "problem" with firms buying existing capital goods. Those who have them now sell them.

Usually what these post Keynesians really fear is that the market clearing real interest rate is negative.

It is a worry that I have as well, but I think nominal interest rates on short and safe assets should go negative in these cases.

I suppose those wedded to a gold standard or a fixed quantity of currency-like base money would have problems. But I don't favor either of those things.

I agree with you that perfect intertemporal coordination is unlikely. What does the recession look like when it money expenditures continue to grow at a stable rate? Less production, lower productivity, shortages of products, and higher prices? Some areas of the economy shrink and others grow? What's new?

The problem lies with the existence of a central bank. The problem is that there is no single demand for money, but a variety of local demands for money. A central bank is a giant blunt instrument. It cannot direct money increases to where money demand is high because it does not have local knowledge. And that's the bottom line. This is the benefit of free banking -- its inherent decentralization. What makes anyone think that if prices and product supply cannot be centrally planned, precisely because of knowledge issues, that money supply can be centrally planned? I know there is a certain sense of "well, what do we do with what we have?" but to me this is like arguing about the right way to supply and distribute goods given central planning. The task is impossible. And only by admitting it is impossible can we come up with real solutions (i.e., market solutions replacing impossible central planning).


There is no single demand for candy either, but a variety of local demands. So what? The candy manufacturer just sells to whatever wholesalers are willing to buy; wholesalers then sell to whatever stores are willing to buy; and stores then sell to whatever individuals are willing to buy. The candy manufacturer doesn't have to know exactly which individuals demand candy and where.

In the case of central banking, the process can take place in reverse, because central banks often announce their intention to purchase assets. In anticipation, primary dealers begin bidding for government bonds before the central bank has moved, thus driving down their interest rates. They purchase first from people who are selling at the lowest price, i.e. people with the highest demand for money, and last from people who are selling at the highest prices, i.e. people with the lowest demand for money. The Fed will then pay the primary dealer however much money is necessary to purchase the bond in turn, because there is always some price at which they would rather have more money than the bond.

By merely announcing they will soon increase the supply of money, central banks can set in motion a series of transactions that satisfy the demand for money before a single OMO has been completed.

So what if the excess demand for money is focused in people who do not have government bonds to sell, how do they get the money? Because money is the medium of exchange. Money is that which we accept in exchange because we expect it can be readily used to buy stuff; there is always some price at which we would rather have more money than the bond. The excess money balances of these individuals will then be spent on goods and services at the lowest prices they can find. The people who sell at the lowest prices demand money more than people charging higher prices. The whole price system then ensures, like the distribution network for candy, that money ends up in the possession of those who demand it most.

The main distortion this creates is just lower interest rates on government debt. Not a good thing, in my opinion, but it is not the same problem you describe.


If you are going to get in an argument about data with me (generally not a good idea), you had better do a good job of it than you have so far, although you have already in advance said you don't believe the data and think FDR was a rotten, unemployment-inducing bum, no matter what the data is.

So, I just googled "US unemployment rates 1930s," and the first source to give a rate for each year was something called "," probably a total fraud. But here are their numbers, which, frankly, fit with what I have seen many times in the past.

1930 8.7%
1931 15.9%
1932 23.6%
1933 24.9%
1934 21.7%
1935 20.1%
1936 16.9%
1937 14.3%
1938 19.0%
1939 17.2%

If you wish to continue claiming that FDR got the rate down to 8% before driving it back up to 15%, I would suggest that you provide at least a remotely credible source, please.

> I agree with you that perfect intertemporal coordination
> is unlikely. What does the recession look like when it
> money expenditures continue to grow at a stable rate?
> Less production, lower productivity, shortages of
> products, and higher prices? Some areas of the economy
> shrink and others grow? What's new?

In that case I don't think we disagree much.

I'm not that worried about the market rate of interest going negative. I think you've been right that this is less of a concern than many think.

What I'm thinking about more is the number of price readjustments that are necessary for real GDP to rise again. Not just the movements needed in large markets like bonds, but the stickier movements needed in
the connected markets that are broadly related to investments. Anyway, I think we pretty much agree about that.


As I'm sure you know there is a big debate over which series is more valid for unemployment in the 30s. Those numbers are the Lebergott series, whereas the Darby series runs:

1933: 20.9
1934: 16.2
1935: 14.4
1936: 10.0
1937: 9.2
1938: 12.5
1939: 11.3

So Troy's 8% is still a little low, but not seriously so for 1937, according to the Darby series.

For those who don't know, the difference between the two series is whether one counts those with jobs in gov't make-work schemes as employed. Darby's does, Lebergott's does not. Which makes more sense probably depends on the point you wish to make: the effect of policy or the behavior of the private sector. For Troy's purposes, if he wants to show the effects of FDR's policies, then the Darby sequence makes more sense perhaps.

If you want to argue data Barkley, you should probably do better than and provide the appropriate context for what your data are actually measuring.

I had a piece published today on the FOMC's recent move.


I was in a hurry, and you admit that the series I provided is more or less the alternative to Lebergott. Ironically, the Lebergott series undercuts Troy's argument more than the one I provided. He was fussing at FDR not for having supposedly gotten the UR down to 8%, but for shooting it back up to 15% or whatever. As it is, the series I showed has nearly that much increase in nominal numbers, from 14.3% to 19% between 37 and 38, whereas the Lebergott one has it only going up from 9.2 to 12.5.

BTW, it occurs to me that that the difference between those two differences (for 37 to 38) may well be a not too far off measure of the impact of cutting back fed spending related to temporary employment, rather than due to the tax increases emphasized by so many, which probably contributed to the 3.3% difference in the Lebergott numbers, whereas the other series has the difference at 6.7%.

Anybody here really want to argue that the fiscal tightening in 1937 did not contribute substantially to that double dip? (BTW, it was that second dip that led to the coining of the term "recession" initially.)

Steve's numbers come closer to my numbers, which I was reciting from memory, but I must admit that Barkley's numbers are far better proof that FDR was a horrible failure as a President. And mostly because he did things Barkley approves of. Here's a lovely chart:

It's from a blog, but the chart itself is a Heritage chart. They point out where the double dip occurs, with the second New Deal being put in place. I will also note that this chart points out that unemployment went down because we sent most of them off to war, but doesn't discuss the detrimental economic effects of that war, which cannot be demonstrated by either unemployment rates or that ridiculous fiction known as GDP.

The comments to this entry are closed.

Our Books