Why did monetary policy fail?

I’ve already argued that the current depression was caused by an excessively tight monetary policy.  But why did policymakers get it so wrong?  I don’t think it’s just the Fed, there is a deeper problem in way the profession as a whole approaches these issues.

Yes, the Fed has made some bad decisions.  But as we saw in the bank bailout case (when the original bad debt purchase plan was replaced by equity injections), policymakers do respond to strong criticism from the economics community.  Unfortunately, I see little evidence that macroeconomists understand how the Fed (and ECB and BOJ) caused the current depression, and thus the profession as a whole is complicit in their policy errors.  At the deepest level, this depression is a attributable to a failure of imagination, a failure to look beyond the surface of current events.

(In using the term ‘depression’ I don’t mean to suggest that we are in another Great Depression.  I hope and expect this downturn will be much milder.  It is already pretty severe, however, and is showing signs of the “debt-deflation” experienced during the Great Depression.)

Of course the policy errors were unintentional, nobody wanted this to happen.  If the errors were inadvertent, then the profession must have misunderstood the nature of the problem, or the effectiveness of potential solutions.  But why were these mistakes made?  I can see at least seven areas where policymakers took a commonsense approach, whereas what was actually needed was a counter-intuitive perspective:

1. Misdiagnosing the problem. The scale of the sub-prime crisis that emerged in August 2007 was initially underestimated, but otherwise the problem was correctly understood as being sector-specific (assuming appropriate monetary policy.)  Unfortunately, many economists failed to see that one year later the nature of the crisis changed suddenly, and dramatically.  What started out as a financial crisis with little impact on nominal GDP, became a massive collapse in aggregate demand after July 2008.  For the first year of the subprime crisis nominal spending held up pretty well (except in residential real estate, which had been declining since mid-2006.)  After July 2008, however, prices and output began falling in a wide range of industries, and as a result the banking crisis intensified.  Like any other other demand-side slump, the root cause was a failure of monetary policy.

Most pundits continue to misdiagnose the problem, mistakenly assuming that we needed to “fix” the financial system in order to generate a recovery.  (One recent exception is Martin Wolf.)  Bank bailouts will not work, however, if nominal spending continues to decline.  Without a shift in monetary policy the banking crisis will only get worse, as falling AD makes more and more borrowers default.  On the other hand if we use monetary policy to boost nominal spending it won’t merely help the macroeconomy, it would also greatly reduce the severity of the banking crisis.

2. Insufficient attention paid to nominal GDP. Most macroeconomists focus on real output and prices, but in an asymmetrical fashion.  To grossly oversimplify, low inflation is usually viewed as a good thing, whereas low output is a bad thing.  But the central bank doesn’t directly control either variable, rather it affects each through the impact of monetary policy on nominal spending.  When there is a precipitous fall in nominal GDP, it is easy to overlook the role of monetary policy if one focuses separately on prices and output.  Thus in the 4th quarter of 2008 nominal GDP fell roughly 5%, almost entirely due to a decline in real GDP.  At first glance the approximate price stability might not seem so bad, after all isn’t price stability the Fed’s goal?  At a recent economic conference I heard a number of conservative economists asking “Deflation, what deflation?”  And falling real output could be due to any number of nonmonetary factors–including the problems in the financial system.

From the perspective of nominal GDP things seem much different.  The Fed generally aims for a long run increase of about 5% in nominal GDP (3% real long run growth (assumed to be independent of monetary policy), and their implicit target of roughly 2% inflation.)  The Fed may not be able to control real GDP, but it can certainly control nominal GDP with an aggressive enough policy (see my earlier proposal here.)  So from this perspective the fact that nominal GDP fell 5% last quarter, and may be falling as much as 10% this quarter, is a flashing red light that monetary policy is far too contractionary to meet the Fed’s implicit spending targets.

3. Reversing causality. Many economists simply assume that the current contraction has been caused by the financial crisis.  After all, isn’t that obvious?  Actually, no.  For nearly a year after the onset of the financial crisis nominal GDP continued growing at better than a 3% clip.  Now it is plunging.  Most seem to assume that this new state of affairs was somehow caused by the Lehman failure, and the subsequent loss of confidence in the entire financial system. My view is that this reverses the causality; it seems much more plausible that the current problems in the financial system are being caused by the recent (and expected future) sharp fall in nominal GDP.

An unexpected decline in nominal GDP puts stress on a banking system that is based on nominal debt.  It is even more disruptive to a system that has become highly leveraged in expectation that the Great Moderation would continue, i.e., that central banks would insure that we never again saw a repeat of the plunging nominal GDP of the early 1930s.  And it is devastating in a system that was both highly leveraged and already buffeted by severe losses in residential lending.

The 2007 subprime crisis did not cause the stock market crash of 2008, as stocks were still only modestly below record levels in early June 2008.  The October crash occurred when investors began to see deflation and depression on the horizon, and saw that loan losses would spread from the already weakened subprime sector into the sort of commercial and industrial loans which would have been sound had nominal GDP continued expanding at close to 5%/year.

4. Assuming that monetary policy has been expansionary. Too many economists merely look at the sharp fall in interest rates, and the sharp increase in the monetary base, and assume policy has been expansionary.  But the Fed also cut rates sharply and increased the base during the early 1930s.  Just as during the Great Contraction, policy has been highly contractionary in the only sense that matters, relative to what is needed to meet the Fed’s policy target for nominal spending.

Nominal interest rates are not a reliable policy indicator.  In another post I pointed out that a highly contractionary policy will generate deflationary expectations, and drive nominal rates toward zero.  And once that occurs the real demand for bank reserves will rise sharply, especially if the Fed foolishly begins paying interest on reserves.

5. Backward-looking monetary policy. Too many economists use the structural modeling approach to policy; looking at the past performance of key variables like inflation and real growth.  But this won’t be good enough when expectations are shifting rapidly from inflation to deflation.  In October 2008 the stock, commodity, and bond markets all saw the oncoming collapse in nominal spending before the Fed, despite the fact that lagging 12 month inflation rates continued to exceed the Fed’s comfort zone.  As a result, the Fed was far too timid in easing policy, indeed it is not clear they eased at all, as the cut in the target fed funds rate and injection of reserves was neutralized by the policy of paying interest on those added reserves.

The commonsense view is that markets are highly irrational, especially the stock market.  This is because economists tend to assume that a sudden large change in equity prices is irrational any time they are not able to identify the fundamental cause.  By October 2008, the financial crisis that occurred in September seemed to ease slightly, and thus commentators attributed the stock market crash to irrational investor fears.  But just as in October 1929, and October 1937, the markets saw a collapse in aggregate demand well before the experts.

If economists fully embraced the need for a forward looking policy, we would have already given up on trying to model the monetary transmission mechanism.  John Cochrane mentioned (in an email) that he was surprised that leading new Keynesians hadn’t latched on to my futures targeting idea; that it was so obvious in retrospect.  Obvious perhaps, but only if one understands that effective policy must be forward-looking.

6. The liquidity trap fallacy. Recent events have demonstrated the degree to which economists’ perceptions are driven by uninformed assumptions.  We teach our students out of textbooks (such as Mishkin’s best-selling money text) that say monetary policy is highly effective even when interest rates fall to zero.  But apparently few economists believe what they teach, as the majority seem to favor fiscal stimulus based on what Keynes would call “voices in the air,” erroneous perceptions of past so-called “liquidity traps.”

Why are points 5 and 6 so important?  Because if monetary policy is always capable of boosting AD, and if policy should be forward-looking, then the Fed has both the ability and duty to make sure that nominal spending is expected to grow at its target rate.  The Fed’s failure to do so after September 2008 was not just a missed opportunity to address the economic downturn, it was the cause of the economic collapse.  If doing X can easily prevent problem Y, then not doing X is the cause of problem Y.

The failure to effectively manage expectations led to a major loss of credibility, as nominal spending is now expected to undershoot the Fed’s implicit target for years to come.  This policy failure makes the Fed’s traditional policy tools even less effective, and makes it even more essential that they adopt unconventional methods.

7.  Irrational fear of hyperinflation. Some of the reluctance to try unconventional monetary stimulus (which almost everyone concedes would work if pursued a l’outrance), is the fear of excessive inflation, of overshooting the target.  Economists see that a large increase in the monetary base has accomplished little, and can only imagine how aggressive policy would have to be in order to turn expectations around.  This fear is partly based on a (mistaken) concern over policy lags.

In fact, an effective policy (such as futures targeting) would probably involve a much smaller increase in the base than what has already occurred.  Once again the danger lies in confusing cause and effect; the huge increase in the real demand for bank reserves is an effect of the deflationary policy (and the interest payments on reserves) not an indicator of policy ease.

Despite the existence of policy lags, there isn’t much risk of monetary stimulus leading to high inflation in the near future.  If the stimulus threatened to overshoot, we would see inflation expectations show up in the indexed bond market, and elsewhere.  The greatest risk we now face is that it will take the Fed too long to recognize that the risks of inaction far exceed the risks of aggressive easing.


Of course most economists suffer from only some of these misconceptions.  I have noticed that left-leaning economists tend to be more sensitive to the problems created by inadequate nominal spending, but often fail to see how monetary policy can address that issue (and thus how the contraction was caused by inadequate monetary policy.)

Those on the right are more likely to acknowledge that monetary policy drives nominal spending, even at the zero rate bound, but some fail to see just how much of the current crisis is driven by falling  nominal expenditure.  Others see the relationship between nominal shocks and business cycles, but assume that the Fed has already taken effective action, and that we merely need to wait for the “policy lags” to play out.  Just why some free market economists ignore the overwhelmingly negative reaction to recent Fed policy in the financial markets, remains a mystery to me.



49 Responses to “Why did monetary policy fail?”

  1. Gravatar of Alex Golubev Alex Golubev
    25. February 2009 at 07:33

    Thank you for sharing your great analysis. I believe that what really got the Scandinavian countries out of their funk was not nationalizing the banks, but devaluing their currencies. During the GD1, the countries that devalued first, started recovery earlier. Devaluations hurt responsible savers, so it’s not necessarily the fairest solution. but who said that saving in the domestic currency vs gold for example, is a good idea? 🙂 The way i see it, devaluation is the only remaining EFFECTIVE (but not necessarily fair) option. i don’t think there are any completely fair options left. What tools to we have to devalue? Issue treasuries and have the Fed buy them in record amounts? what about simply changing the reserve requirement for banks?

  2. Gravatar of Un nouveau blog de macro monétaire bigrement intéressant « Rationalité Limitée Un nouveau blog de macro monétaire bigrement intéressant « Rationalité Limitée
    25. February 2009 at 09:01

    […] point de vue d’un macroéconomiste manifestement non-keynésien… mais qui rejette aussi l’explication autrichienne […]

  3. Gravatar of Gaurav Gaurav
    25. February 2009 at 09:31

    Testing a comment

  4. Gravatar of Jon Jon
    25. February 2009 at 10:01

    The Fed has been sterilizing its monetary intervention quite substantially. This was explicit policy throughout 2008–first by selling down their stock of treasuries and more recently by having the Treasury sell debt and then deposit the money at the Fed.

    The contrast is quite stunning compared to the growth in the Fed’s balance sheet. More recently, I believe the sterilization has been accidental. The Treasury’s general account at the Fed has ballooned–reflecting that the treasury is selling debt well before it disperses TARP funds.

    If you subtract out the bank reserves, the monetary base declined in the 3rd quarter of 2008! Definitely a policy of contraction. Only just recently did the Monetary Base less reserves reach trend again.

    Since September there has been a substantial decline in domestic high-powered money. I estimate close to 30% based on the H.4.1 Fed report. This is due to the Treasury’s cash balances (20x normal), bloated bank reserves, Treasury transfers from the bond market into Bank Capital (which is itself sterilized money), and foreign currency swaps by which we’ve been saving the European banking system.

  5. Gravatar of James Hanley James Hanley
    25. February 2009 at 11:09

    Great blog. Now I have to add yet another econ blog to my list of required reading!

    I have a question, which I’m not a good enough economist to answer, so I dare to express my hope that you will. I have seen it said that monetary policy has reached its limit, as interest rates are near zero. This seems illogical to me, as zero is not a necessary stopping point. Clearly it is mathematically possible to have negative interest rates (that is, pay people to borrow money).

    So I’d love to have a monetary theorist’s take on the idea of using monetary policy so aggressively–in the current crisis–that it moves into negative interest rate territory. Would such a policy likely be effective? Would it have drawbacks unique to being in negative rate territory?

  6. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    25. February 2009 at 12:18

    Am I understanding you correctly to be saying that Milton Friedman’s dislike of monetary policy operating through a focus on interest rates was well founded?

  7. Gravatar of Bill Woolsey Bill Woolsey
    25. February 2009 at 14:58

    James Hanley,

    The problem with negative interest rates is that the financial system is based upon zero interest hand-to-hand currency. No one will lend at negative interest rates if they can hold zero-interest currency instead. I think it is the cost of storing currency (like in a bank vault or safety deposit box) that puts the lower bound on interest rates–which would be lower than zero.

  8. Gravatar of Alex Golubev Alex Golubev
    25. February 2009 at 15:08

    Bill Woolsey,
    i think the point is that the Fed oughta go out there and lend at negative rates to stimulate borrowing. the questions is about that option. I never really understood why 0% would be a limit myself. if the Fed wants to create money, then why not pay interest to incentivize borrowing?

    James Hanley,
    In my (stupid) opinion, that part of the equation is immaterial. they’re better off trying to get CREDIT spread part of the rate equation under control by creating agencies like RFC or whatever to bypass banks and lend directly to business/consumers.

  9. Gravatar of ssumner ssumner
    25. February 2009 at 17:29

    Alex, I agree that devaluation is a good solution for a single country (Japan circa 1998), but when the entire world is facing falling AD you need all currencies to be worth less. The only way this can be done is to have a monetary policy expansionary enough to reduce all currency values, not against each other, but against goods and services. Have all currencies depreciate at 2% a year against goods and services. That means (mild) inflation

    Jon, I agree that a sort of sterilization occurred, but the monetary base did go up sharply after September (check St. Louis Fed website.) In my view what caused sterilization was the decision to pay interest on reserves. This turned reserves into a sort of T-bill-type investment. Even better in fact, as the interest rate (although low) was higher than the yield on T-bills. So the Fed did increase the base sharply, but it didn’t increase the non-interest-bearing base sharply.

    James, Bill is right that you can’t really have negative interest on money. But I do think you can have negative interest on a part of the monetary base–bank reserves. The Fed could charge a 1% penalty on excess reserves, to encourage banks to hoard less reserves. I think Bill has also been working on this topic.

    Patrick, The answer is yes.

    Alex, See earlier response. Also, wouldn’t banks just arbitrage the difference? They’d borrow from the Fed and simply hoard reserves–earning a risk free profit. I’d have to think about this more, but right now I’m skeptical.

  10. Gravatar of Jon Jon
    25. February 2009 at 19:31

    Unfortunately my site (lostdollars.org) is down, but I do weekly fetches of all of the statistical releases and do some automated analysis. You are correct that the monetary base has gone up, you are also correct that its only the interest bearing faction. Once you subtract out reserves, the monetary base less reserves has been below trend throughout the crisis until just recently.

  11. Gravatar of What Guru’s are Saying : Readings 02/26/09 What Guru’s are Saying : Readings 02/26/09
    26. February 2009 at 00:56

    […] Why did monetary policy fail? (TheMoneyIllusion) […]

  12. Gravatar of Readings 02/26/09 | Venture Capital Bloggers Network Readings 02/26/09 | Venture Capital Bloggers Network
    26. February 2009 at 02:50

    […] Why did monetary policy fail? (TheMoneyIllusion) […]

  13. Gravatar of Bill Woolsey Bill Woolsey
    26. February 2009 at 04:27

    As Scott says, the key to negative interest rates is for the Fed to charge banks for holding balances in their reserve accounts. The Fed could lend to banks at a negative interest rate and then charge them for holding the reserves. Similarly, there is nothing impossible about the Fed offering to pay more than face value of T-bills, giving them negative yields. In such a scenario, banks would have to charge people for holding deposits.. and so on.

    It all breaks down because banks can withdraw currency from reserve accounts and keep it in the vault. People can withdraw currency from banks and keep it in safe deposit boxes.

    I suppose everyone just holds currency and the Fed holds all the other assets.

    It only works in system without currency or where currency payments have been suspended.

  14. Gravatar of GG GG
    26. February 2009 at 04:53

    This is precisely why academic economists should have zero say in the causes/fixes of the current crisis. Nominal GDP does not appear out of thin air, but is the result of production and confidence. I know it’s hard for you guys to think in rational ways.

    This recession is 100% related to the financial crisis, because it was easy credit that led to the overexpansion of the economies in the last decade. Monetary policy had a role, but nothing ruled greater than the ongoing implementation of Basel II and the perverse incentive by the financials to structure AAA debt to keep lending more and stoking the economy. Nothing more, nothing less.

    The reason that a subprime mortgage crisis turned into a global contagion is that few people appreciated in August 2007 just how many of the MBS were sitting on the balance sheets of the financials, as opposed to the conventional wisdom of being parceled out to hedge funds and other investors.

    The trouble on the horizon for the C&I loans and debt is not because the economy is shrinking, but because you’ve eliminated at least 30% of investor demand for all the leverage that was put on by the summer of 2007.

    Many companies that will default will do so not because they can’t keep paying interest, but because they won’t be able to refinance debt that’s set to come due in 2010 & 2011. For that reason, many banks will start pulling the plug in 2009 to preserve distressed values.

  15. Gravatar of ??? ???
    26. February 2009 at 06:04

    Brilliant analysis. However, one question: You make a very good about the nominal GDP dropping and being a main factor in the current depression. But what caused it to drop?

  16. Gravatar of Chris Chris
    26. February 2009 at 06:13

    I’m pretty close to economic illiteracy, but I’ll risk a question. I think I understand how encouraging borrowing will boost AD here and help us out. But Martin Wolf was mentioned above, and isn’t he right in his analysis of debt imbalances from a global perspective? Doesn’t that imbalance have implications for the long term effectiveness of monetary policy? I can see how this would work in the short run, but isn’t it a recipe for ever increasing consumer debt? Doesn’t that have a finite if unknown limit? Please be kind to the neophyte.

  17. Gravatar of Carl Futia Carl Futia
    26. February 2009 at 07:57


    You suggest that the fall in nominal AD caused the banking crisis. But what caused the fall in AD? Was it simply that the fall in construction spending coupled with the wealth effect arising from the drop in housing prices became too big to be contained?

    My own guess is that we are experiencing the consequences of an “uncertainty shock” to AD. This shock reached its “tipping point” when Lehman Bros. went under.

    In late September, just before the House of Reps rejected the first TARP proposal, the S&P 500 was trading around the 1220 level, down about 23% from its 2007 high 11 months earlier. But once the House rejected the first TARP, the average proceeded to fall another 32% in less than 2 weeks!
    It was at this point I think that recessionary and deflationary expectations took hold with a vengeance. And the reason for the change in expectations was widespread fear about the solvency of the banking (and near-banking) system. This generated the “uncertainty shock” to AD.


  18. Gravatar of John Maxwell John Maxwell
    26. February 2009 at 08:41

    According to the Case-Shiller home price index, housing prices have dropped to 3/4 of their peak. In order to bring nominal house prices back to the peak level, you would need to inflate the money supply by 4/3 (33%). That would cause tremendous disruption to the rest of the economy — I doubt that the Fed is willing to go that far. On the other hand, if there was an *expectation* of 10% inflation per year for the next few years, that might solve the problem. How much do you think that the Fed needs to inflate in order to right the economy?

  19. Gravatar of Plamus Plamus
    26. February 2009 at 11:07

    “Despite the existence of policy lags, there isn’t much risk of monetary stimulus leading to high inflation in the near future.”

    Why do you feel the need to hedge this bold statement by qualifying “in the near future”? Are we all dead in the long run, or do you feel that we have a fire to fight, and should not care about dumping the house into the river?

  20. Gravatar of Greg Ransom Greg Ransom
    26. February 2009 at 15:46

    A boom-bust cycle is often geographic in nature.

    Orange County, CA was already in steep recession in 2007, with structural sectors leading the bust: construction, finance, real estate, etc.

  21. Gravatar of ssumner ssumner
    26. February 2009 at 17:28

    Bill, Don’t banks have to report vault cash data to the Fed? They are counted as a part of reserves. If the Fed has no data on vault cash, then my negative interest in reserves proposal won’t work. I intended to also penalize vault cash.

    ??? See my comment near the end of “My 15 minutes” Comment 16, part addressed to Robin.

    Chris: Don’t worry about being a neophyte. Your question isn’t easy to answer quickly. There are macro issues (total spending) and micro issues (individual people or industries that got overextended.) I don’t favor bailing out specific industries. Three percent inflation won’t make the sub-prime crisis disappear, but it will support the broader economy–as the broader economy was holding up pretty well in the first 6 months of 2008.

    Macro is not a zero sum game–when you have good macro policy then total output in the economy is greater, and the sum total of micro problems (like debt defaults) is smaller.

    Regarding what caused AD to fall, see my comment toward end of “My 15 minutes” as addressed to Robin. On your other point, industrial production began its sharp fall in August, that was before Lehman. Certainly causation went both ways, but I think its most useful to think in terms of Fed errors of omission causing lower AD.

    John, 2% or 3% inflation won’t save the bad sub-prime housing loans, nor do I favor fast enough inflation to make bad loans good. Rather I favor enough inflation to stabilize the macroeconomy, and that will prevent the home loan crisis from spreading quite as deeply into non-subprime loans. But again, monetary policy should not focus on particular sectors, it should focus on nominal GDP (or inflation.)

    Plamus, I shouldn’t have hedged. I intended no implication that high inflation is coming later.

    Greg, I agree about regional recessions. I don’t see regional recessions as the proper focus of monetary stabilization policy. The are part of the market process of creative destruction. Monetary policy should focus on national variables. Is that also your view?

  22. Gravatar of ssumner ssumner
    26. February 2009 at 17:31

    I forget GG, All I can say is that had the financial crisis not occurred, the recession would not have occurred–as you say. But had the Fed kept nominal GDP growing at 5% a year, it would have been a mild “stagflation-type” recession without nearly as large budget deficits, bailouts, nationalization, etc.

  23. Gravatar of Bill Woolsey Bill Woolsey
    27. February 2009 at 02:44


    The Fed reports vault cash used to meet reserve requirements, so I presume some reporting is going on. So, your notion is that the Fed could continue to charge banks for how much vault cash they hold. While they would have an incentive to lie about it then, I suppose this is now worse than their usual incentive to claim more vault cash than they have to meet reserve requirements.

    So, that leaves the entire problem to be currency drain from the banking system and storage in safe deposit boxes and the like.

  24. Gravatar of Bill Woolsey Bill Woolsey
    27. February 2009 at 03:24

    Yeager has an paper called “Money and Credit Still Confused.”

    I am sure that most everyone here receives some sort of income and makes expenditures. I expect that nearly everyone here has the income appear in your cash balance–checking account or currency holdlings. And then you spend it and it leaves. Money flows through your cash balance.

    If your cash balance is larger than what you want to hold, you spend it. However, if your balance is too low, you cut your spending. What you do or don’t spend on isn’t important to the argument. You change current expenditures on things other than money to adjust your actual money holdings to your desired money holdings. While you may be paying down debts or buying securities when you spend money, and, of course, you can borrow money and spent it too, all of that is irrelevant to the basic argument.

    The total money that has been created is currently held by someone. This is the liability side of the Fed’s and the banks’ balance sheet.

    It is true, of course, that the banks and the Fed lend. That is the asset side of their balance sheets. But the supply of money is the liability side.

    If the supply of money created is greater than the amount people want to hold, they they spend more. If the amount of money is less than the amount people want to hold, then they spend less.

    This is how money creation is “supposed” to manipulate total spending in the economy. Not through getting people to borrow and spend. Leaving aside this money creation, with a credit transaction, the borrower spends more, but the lender spends less. When debt is repaid, the former borrower spends less, but the former lender spends more. It is, of course, possible that those receiving debt repayments will just hold the money received. It is possible that people will lend money they would just hold.

    And then, we are back to the process described above. What happens when people want to increase their money holdings. They receive money payments and don’t spend them.

    For the supply and demand for money process to NOT work, it requires that somehow the creation of money causes a matching increase in the amount of money people want to hold. (And a decrease in the quantity of money cause a matching decrease in the amount of money people want to hold.)

    It doesn’t require that people always want to hold the same amount of money. That it could change, and cause problems, is a key insight of this approach. The “policy” implication is that if the money supply changes to accomodate changes in the demand for money, then total spending will not be depressed (or expanded) by a change in the amount of money people hold. Sumner has said several times in this thread that such is the “job” of the Fed. If there is an increase in the demand for money, FOR ANY REASON, the Fed’s job is to raise the quantity of money to accomodate it. Sumner is describing a fall in expenditure due to a failure of the Fed to do what he considers its job to be “caused” by the Fed. He isn’t claiming that everything would have been just fine and then the Fed suddenly took some wrongheaded action that started the problems. It is that the Fed failed to do its job of accomodating changes in the demand for money.

    If the demand for money rises because of a loss in confidence (less desire to hold other kinds of assets because of perceived risk,) then the Fed should accomodate it.

    The only way this won’t work is if when the Fed increases the quantity of money, that in and of itself, causes the demand for money to raise as much (or worse, I guess, more) than the increase in the quantity of money.

    That is what the “liquidty trap” is all about.

    Increasing the quantity of money doesn’t create excess money balances for individual firms and households on net, because they just want to hold more money.

    Some of the semi-Keynesians as well as people noneconomists who are focused on debt (they seem a bit like pre-Keynesian liquidationists) either don’t understand this, or they believe that their is some kind of liquidity trap.

    On the other hand, one of the ways that an imbalance between the quantity of money and the demand to hold it can be corrected is through a change in the price level, including nominal incomes like wages. You can either describe this as the real quantity of money ajusting to the demand or else the nominal demand adjusting to the supply. Regardless, once the supply and demand for money is balanced, real expenditures adjust to the proper level.

    The liquidty trap supposedly can keep this from working too. Sumner (and I) would believe that it can work, but that it is very disruptive and so a bad idea.

    Anyway, there are some new classical macroeconomists and many “Austrians” who believe that this process works more or less perfectly. And so, total real expenditures is pretty much always where it should be. What Sumner (and I) would say that the Fed should do, is pointless. There is no problem to solve.

    From their perspective, the only “problems” that we might have disruptions in real financial intermediation. Or, perhaps, imbalances in the production of particular types of goods.

    I believe it is difficult to imagine that there are not problems in real intermediation and imbalances in the types of goods being produced. And I don’t believe that Fed policy should be aimed at fixing them. But there is another problem going on now. And that is too little spending, which must be caused by an imbalance between the quantity of money and the demand to hold that money. And it can be fixed by the Fed.

    To the degree that shifts in composition of demand and problems in real intermediation result in disruptions in our abilities to produce goods, then keeping nominal income on target will not prevent production from falling (or growing more slowly) and it will result in rising prices and lower real wages. For a time.

    But there will not be this phenonenon of reduced spending, sales and production of goods pretty much across the board. And that is the problem that the Fed can and should solve.

  25. Gravatar of John Maxwell John Maxwell
    27. February 2009 at 08:20


    In order for housing prices to revert to the mean, they have to fall 40-50%. This is going to affect most loans made in the last few years, whether or not they were subprime. People aren’t going to be willing to make mortgage payments that are twice what they should be, even if they don’t lose their jobs. So just fixing the economy isn’t enough to save the banks.

    I think that the underlying cause of the current crisis is that the US credit market debt had grown to 350% of GDP. Over the last 100 years, the typical amount of debt was 170%. It was 260% right before the Great Depression.

  26. Gravatar of Bill Woolsey Bill Woolsey
    27. February 2009 at 13:00

    Why does excessive debt cause a crisis? What kind of crisis does it cause?

  27. Gravatar of GG GG
    27. February 2009 at 13:41


    I definitely agree with that, and that’s what the markets were preparing for when everyone recognized that real estate had gotten out of hand in 2006.

    Monetary policy is secondary to getting to the bottom of bank balance sheets.

  28. Gravatar of ssumner ssumner
    28. February 2009 at 12:10

    John, I agree that easy money cannot solve the sub-prime part of the banking crisis. That crisis was manageable until mid-2008. What happened after mid-2008 is that the highly deflationary policies of the Fed (again by accident on their part) have made loans that would otherwise be good turn bad. At least a portion of higher quality home loans are deteriorating precisely because of the sharp fall in nominal GDP. And a huge part of their deteriorating industrial and commercial loan portfolios are due to the fall in NGDP.

    Bill, The Yeager stuff on the distinction between money and credit is very good, and I am planning a post for our non-economist readers on the excess cash balance transmission mechanism–as the press usually just discusses monetary policy in interest rate/credit terms.

    Regarding your preceding comment, you know we are in trouble when we are discussing vault cash and safety deposit boxes. (shades of 1933) I’m just thinking out loud here, but if Fed policy did succeed in reducing excess reserves with a penalty rate, and if the base money spilled out into cash in circulation, isn’t it unlikely that it would all be hoarded as a T-bill substitute? If it would, why wouldn’t massive quantities of cash already be hoarded. In the Great Depression currency hoarding occurred before any liquidity trap, and was (partially) accommodated by the Fed. Perhaps the answer to my question is that T-bill yields would fall slightly negative, just enough to reflect the cost of safety deposit boxes. Anyway, it seems worth a try before getting into quantitative easing.
    I have studied liquidity traps for years but they still confuse me. I sometimes think that monetary policy in a liquidity trap is the hardest thing to understand short of quantum mechanics, or what is consciousness, or why is there something rather than nothing?

  29. Gravatar of John Maxwell John Maxwell
    28. February 2009 at 20:35


    I’m only an amateur, but I will try to explain why I think that excessive debt can lead to a crisis. When an individual takes on too much debt, they reach a point where they realize that they need to start paying down their debt. Usually they do this by reducing consumption and selling assets. If a large number of individuals do this at the same time, then you can get a dramatic reduction in consumption along with reduced asset prices (because there are more sellers than buyers). You also get people defaulting on their debt, which reduces the money supply. This process can feed on itself, with the reduction in the money supply and the decrease in the velocity of money causing further asset declines, which cause people to increase their savings by decreasing their consumption so that they can make up for the loss of wealth caused by the asset reductions.

    In the current crisis, I think that there was a cascade of problems. Hedge funds that had borrowed too much realized that they needed to sell assets to cover their loans. Then banks that had too much exposure to hedge funds or to subprime mortgages started to increase their reserves to cover against future losses. The October stock market crash was caused when investors decided that the authorities weren’t going to be able to contain the crisis. The housing crash and the stock market crash convinced consumers that they needed to spend less and save more. This caused companies to cut back expenses and hoard cash. Then the whole process fed on itself, since banks became worried that companies wouldn’t be able to meet their loan payments. And so on.

    I like Scott’s approach to the problem. I think that the Fed needs to establish an expectation of inflation to get the economy going again. But I don’t think that it will solve all of our problems.

  30. Gravatar of kball kball
    1. March 2009 at 14:39

    Hi Scott,

    I’m also very much an amateur in economics, and while I tend to lean more leftward, I’ve very much enjoyed reading your blog to date. I was wondering if you could elaborate a little bit more about your belief that the housing crisis was manageable, with the financial crisis at least partially separate and being creditable primarily to the Fed’s recent mistakes.

    A few points that are making me wonder are:

    1) Having been reading Calculated Risk for the last two years, its been evident for a while that the initial subprime problems in 2007 were only the tip of the iceberg in regards to overextended homeowners and unpayable mortgages. Most of the subprime problems came from adjustable mortgages that had initial teaser rates adjusting to a much less affordable rate after some time. During a period of rising home prices, one could refinance out of the adjusted rate, but ones home prices stalled these became toxic.

    The subprime adjustable mortgage rate resets didn’t peak until late 2008, while Alt-A and other option adjustable mortgage rate resets have yet to peak.

    2) The corporate loans that have been having the most problems to date have been those to the financial industry and the construction industry. Over the last few months, there have been declines in profitability and employment across all industries, but bankruptcies and defaulting debt still seem to be focused in construction and finance. To me, this seems to be continued momentum from the debt/housing bubble and ongoing bust.

    I feel like there may be ways to explain these from a monetary perspective, but I currently don’t see them. I’m even more interested in the ways monetary policy could be used to alleviate the damage. Thanks!

  31. Gravatar of ssumner ssumner
    1. March 2009 at 18:13

    kball, You make a good point and I confess to not being an expert on the banking system, but I still feel that people underestimate the importance of the difference between nominal GDP falling at 6.2% and rising at 5%. I wouldn’t argue that it would solve all our problems, but I also believe that it would have a lot of ripple effects that people overlook. For instance, it wouldn’t just mean a higher price of goods and services, it would have a dramatic effect on asset prices, (which tend to fluctuate much more strongly than consume prices.) When I say asset prices I include everything from common stock, to houses, to commercial real estate, commodities, etc. These increases, if large, would have a major impact on reducing debt defaults. I cannot emphasize enough the importance of not just looking at specific sectors of the economy, and assuming that the picture you see is same as would occur with faster nominal growth. Many sectors would do much better. But again, I would never deny that even with my policy we are looking at large losses in banking and real estate. But I still think the huge acceleration in loan losses since August 2008 is mostly the result of a worsening macro environment, and not just a delayed reaction to the sub-prime mess. I may be wrong about banking, but even if I were, I’d still support a more expansionary policy on unemployment grounds. (To summarize my long-winded answer, you probably know more than me about the actual situation in banking, and I probably know more about how subtle the effects on monetary policy are.)

    John, I didn’t answer your comment, as it was addressed to Bill, but you might also be interested in my reply to kball.

  32. Gravatar of kball kball
    1. March 2009 at 20:32

    Hi Scott,

    Thanks for the response. I think you are right that changing the nominal gdp, even if it were entirely via inflation, would do quite a bit for reducing the problems of debt burden. One thing that your answer made me think about was to what extent the distribution of that gdp growth across industries matters.

    I think that one of the reasons we have such a problem in the housing and banking industries is due not to the nominal amounts of money that had been directed into them, but the distortion created by flooding so much money into those sectors without similar increases in other areas. The availability of extremely cheap credit specifically for residential mortgages led to a huge overinvestment in housing, and a great imbalance between the cost of a house and the wages of most people.

    Correcting this distortion seems to imply either a deflation in housing or an inflation in other sectors. You’re almost certainly correct that an inflation in other sectors, if evenly distributed, would be less harmful. But I don’t know how to make that happen; it seems that after the tech crash in 2001, the Fed attempted to use exactly this technique to help the economy recover, and it resulted in the housing bubble, which is now crashing with even more disastrous effects.

    Am I misinterpreting this? Or is it possible to prevent this sort of distortion? Thanks

  33. Gravatar of Lorenzo (from downunder) Lorenzo (from downunder)
    1. March 2009 at 21:16

    Bill Woolsey’s long comment (and this is no criticism) encapsulates why I find thinking about money so difficult. There is the cash you hold, there is the money you spend. The cash you hold is a mixture of ballast (against sudden emergencies) and delayed spending (things, generally largish, to be purchased in future). The money you spend moves around the economy if others also spend it (as they mostly do). So money seems both “static” and “mobile”. But if your cash is in the bank, then much of it is “in motion” too.

    So there is the demand for money-as-cash to be held and the demand for money-for-transactions to be used. So, monetary economics IS like quantum mechanics (something being “both” a wave and a particle)!

    Then there is talking of the “velocity” of money as if the movement of something which flows in lots of joinings and partings by being “grabbed” and “pushed” in various swappings (i.e. purposive exchanges) is like a ball bouncing around. It all makes my brain hurt.

  34. Gravatar of James Hanley James Hanley
    2. March 2009 at 13:27

    Bill, Alex and Scott Sumner,

    Thank you for your responses to my question. I hadn’t thought about the way the banks would act if the Fed was paying them to borrow money. A seemingly simple scheme brought down by rational behavior!

  35. Gravatar of ssumner ssumner
    2. March 2009 at 17:15

    Don’t have much time tonight with all the Krugman stuff, I’ll just say in response to kball that I’d guess the housing crisis was 20% monetary policy, 20% Fannie and Freddie, and 60% bad luck (by which I mean we had the bad luck to suffer from forecasting mistakes by banks, etc.) But I wouldn’t blame anyone for being unconvinced by my guess.


  36. Gravatar of Citizen Economists » What Will it Take to Get Sustainable Recovery? Citizen Economists » What Will it Take to Get Sustainable Recovery?
    30. March 2009 at 07:09

    […] I particularly liked the following posts on Sumner’s blog: Why did monetary policy fail?; and The Economics […]

  37. Gravatar of Bela Bela
    3. April 2009 at 10:19

    Greeting. Be careful that victories do not carry the seed of future defeats.
    I am from Luxembourg and learning to write in English, give true I wrote the following sentence: “Newly launched web site farecast uses statistical information to help you decide when you should buy the tickets for your.”

    Thank you very much :(. Bela.

  38. Gravatar of TheMoneyIllusion » Welcome Bloggingheads.tv viewers TheMoneyIllusion » Welcome Bloggingheads.tv viewers
    5. April 2009 at 11:20

    […] For a view of what went wrong with monetary policy look here. […]

  39. Gravatar of The Fed’s Monetary Policy Has Veered Off Course – CBS MoneyWatch.com The Fed’s Monetary Policy Has Veered Off Course - CBS MoneyWatch.com
    30. June 2009 at 04:07

    […] once they had seen that they had already veered off course. The Fed must pay more attention to forward-looking indicators if they want to avoid repeating the costly mistakes of last […]

  40. Gravatar of TheMoneyIllusion » Six reasons to abolish inflation TheMoneyIllusion » Six reasons to abolish inflation
    1. August 2009 at 14:36

    […] post.  If you want a serious look at my policy views please check out my FAQs.  In addition, this link discusses what went wrong last year.  And finally, I want to thank Tyler Cowen for his very kind […]

  41. Gravatar of The 4% Club Is Now Taking Nominations «  Modeled Behavior The 4% Club Is Now Taking Nominations «  Modeled Behavior
    16. August 2010 at 09:26

    […] Scott Sumner Follow Modeled Behavior on Twitter […]

  42. Gravatar of Ist die EZB schuld? « Kantoos Economics Ist die EZB schuld? « Kantoos Economics
    9. November 2010 at 13:18

    […] Eintrag ist inspiriert von Scott Sumner, der seit Anfang 2009 schreibt: die Fed ist schuld! Aber der Reihe […]

  43. Gravatar of だから言っただろうとは言いたくない by Scott Sumner  – 道草 だから言っただろうとは言いたくない by Scott Sumner  – 道草
    24. December 2010 at 22:50

    […] by Scott Sumner  // スコット・サムナーのブログ、TheMoneyIllusion […]

  44. Gravatar of フリードマンの4%ルール(Version 2.0) by Scott Sumner – 道草 フリードマンの4%ルール(Version 2.0) by Scott Sumner – 道草
    24. December 2010 at 22:53

    […] 2.0) by Scott Sumner // スコット・サムナーのブログ、TheMoneyIllusion […]

  45. Gravatar of 金融政策が失敗した理由 by Scott Sumner   – 道草 金融政策が失敗した理由 by Scott Sumner   – 道草
    24. December 2010 at 23:13

    […] by Scott Sumner   // スコット・サムナーのブログ、TheMoneyIllusion […]

  46. Gravatar of ヒュームとコールがFedに提言 by Scott Sumner – 道草 ヒュームとコールがFedに提言 by Scott Sumner – 道草
    27. December 2010 at 07:36

    […] Scott Sumner // 引き続き、スコット・サムナーのブログ、TheMoneyIllusion の初期(2009å¹´2月くらい)のエントリご紹介。今回は、Hume and Hall call out […]

  47. Gravatar of わがロールモデル、ジョージ・ウォーレン by Scott Sumner – 道草 わがロールモデル、ジョージ・ウォーレン by Scott Sumner – 道草
    8. January 2011 at 01:18

    […] わがロールモデル、ジョージ・ウォーレン by Scott Sumner //  ã‚¹ã‚³ãƒƒãƒˆãƒ»ã‚µãƒ ãƒŠãƒ¼ã®ãƒ–ログ、TheMoneyIllusion の初期(2009å¹´2月くらい)の重要エントリ(本人がそうおっしゃってます)をぼちぼち紹介しています。今回は、My Role Model, George Warren(February 28th, 2009)を紹介します。日付に注意願います。サムナー先生の人物像がかなりわかってきます。翻訳はayakkaさんに手伝っていただきました(感謝)。 […]

  48. Gravatar of あなたは合理的期待を「信じて」いるのか?(重要) by Scott Sumner – 道草 あなたは合理的期待を「信じて」いるのか?(重要) by Scott Sumner – 道草
    9. January 2011 at 00:59

    […]  ã‚¹ã‚³ãƒƒãƒˆãƒ»ã‚µãƒ ãƒŠãƒ¼ã®ãƒ–ログ、TheMoneyIllusion の初期(2009å¹´2月くらい)の重要エントリをぼちぼち紹介しています。今回は、自身で「important」と言うDo you “believe” in rational expectations? (important)(February 8th, 2009)を紹介します。  […]

  49. Gravatar of 金融理論の短期講習 by Scott Sumner   – 道草 金融理論の短期講習 by Scott Sumner   – 道草
    11. January 2011 at 01:46

    […] by Scott Sumner   // スコット・サムナーのブログ、TheMoneyIllusion […]

Leave a Reply