The Fed should create the mother of all stock bubbles, permanently.

Before I begin I’d like to thank Dilip, who sends me many useful articles, and Scott (aka “Lawton”) who recommended that I set up the “FAQs” link.  It already got a nice review from Bryan Caplan, so I think it was an excellent idea.  I plan to improve it when I have a bit more time.  BTW, that language snob Bob Murphy insisted “Frequently Asked Questions” should be “FAQ.”  He doesn’t realize that not only are there lots of questions, but they are asked over and over again by newcomers.  So the double plural is required.

No one commented on my pathetic attempt to channel Lovecraft in the final paragraph of my “Mind Snatchers” post.  I’ve always been fascinated by Lovecraft, despite his schlocky writing style.  Especially his collection of letters, which depict a life that mine is increasingly resembling.

[BTW, Michel Houellebecq insists Lovecraft is a superb stylist.  Yes, he’s French, but he makes an interesting argument.]

Speaking of Lovecraft’s style, I am afraid my Austrian readers may find this essay to be an unspeakable horror.  But if you find yourself muttering “Sumner can be provocative, but this time he’s gone too far,” remember that new insights can often come out of very far-fetched thought experiments.

In my Snatchers post I linked to an interesting piece by Leigh Caldwell.  He discussed a speech where New York Fed president Dudley recommended that the Fed try to prevent asset bubbles.  I was once offered a job at the NY Fed, and so I tried to picture myself working in the economics research unit.  I can just imagine us getting a memo from the Fed president, instructing us to figure out when stock prices are inflated, and then recommend corrective action.

When I actually read the speech, it was nowhere near as bad as I had feared.  Dudley did not recommend that the Fed use monetary policy to try to prick stock bubbles, rather he suggested tighter regulations and focused more on housing bubbles.  I don’t think that would work, but at least it might not do much harm.  Nevertheless, it might be worth discussing whether Fed policy could control asset price bubbles, since many others have blamed Greenspan for not popping the tech and housing bubbles.

First we need to consider a few principles:

1.  The Fed is never passive.  They are always targeting something.  It might be the money supply, or interest rates, or inflation, or inflation expectations, or the exchange rate, etc.  But there is never a time when the Fed is doing nothing, and can turn monetary policy over to “bubble fighting” without abandoning its other target.

2.  Monetary policy can only focus on one target at a time.

3.  The Fed cannot prick stock bubbles by raising interest rates; they can only do so by impacting the macro economy.  In 1928-29 they raised rates several times, but were unable to slow the stock market boom.  Only when they raised rates high enough to plunge the economy into a recession, did they stop the stock boom.  In 1936-37 they raised reserve requirements to try to slow down the commodity price boom.  At first it didn’t work, and commodity prices only fell when the Fed pushed the economy into a recession.  Not surprisingly, the Fed has avoided trying to micromanage asset price bubbles after the unfortunate experience of the 1930s.

In my view the basic problem here is that people lose sight of the underlying causes of stock market booms. They don’t occur randomly, but rather tend to occur when the overall macro economy is doing well; when there is steady growth and low inflation.  The late 1920s, the mid 1960s, the late 1980s, the late 1990s, and 2005-07.  The two biggest stock booms (1928-29 and 1999-2000) correspond to what are arguably the two best periods in American macro history.

The Fed can kill off these bubbles if it is determined enough, but why would it want to do so?  If the stock bubbles reflect good economic times, then they can only be stopped by producing bad economic times.  And that doesn’t seem very sensible.  Instead, why not try to make the good times last forever?  I define good times as “5% expected NGDP growth.”  The term ‘expected’ is important, as stock investors are forward-looking.  As we saw in late 1929, stock prices react much more strongly to changes in growth expected over the next 12 months, as compared to the past 12 months.  As long as stable nominal growth with low inflation is expected, then stock prices should be able to stay at relatively high levels—Irving Fisher’s famous “permanently high plateau.”

Just to be clear, I understand that other factor like tax rates also affect stock prices.  If we adopted statist policies then stocks won’t do very well even with stable non-inflationary growth.  But historically it seems like the biggest stock market crashes are associated with expectations of slow growth or high inflation (1987 is an exception.)

Some might argue that my idea is crazy; we have never gone 10 years without a recession, what makes me think we can cure the business cycle?  One answer is that we have gradually lengthened the cycle, as monetary policy has gotten better.  Why rest on our laurels?  Another answer is that Australia just had a 17 year expansion.  If the Aussies can do it, why can’t we?  And another answer is that I am not trying to abolish the business cycle (although I’d like to) but rather I am trying to abolish the expected business cycle.  I am trying to abolish periods like October 2008, when 12-month forward nominal GDP growth was expected to be zero or negative.  If the stock investors can always anticipate solid NGDP growth, that is one less thing for them to have to worry about.  In addition, it would also make financial crises much less severe, as financial crises are made much worse by deflation (recall 1931 and late 2008.).

To summarize, the Fed wants a healthy economy.  Stock investors want a healthy economy.  Let’s give stock investors what they want, with a monetary policy aimed at creating a permanent bubble, i.e. P/E ratios permanently above the historical norm.  After all, the historical norm reflects lots of really, really bad policy.  Why should we settle for an average that includes those sorts of P/E ratios?  Surely no one can disagree with my view that the Fed should do what it can to create permanent (expected) prosperity.  If it leads to high stock prices, so much the better.

Irving Fisher was wrong about the permanently high plateau of stock prices, but he would have been right about the stock market if the Fed had followed his policy advice, and stabilized the price level as Governor Strong did in the 1920s.

PS.  Please don’t tell me that a bubble is temporary by definition.  I obviously mean stock prices too high according to Shiller’s definition of bubbles.


Tags:

 
 
 

90 Responses to “The Fed should create the mother of all stock bubbles, permanently.”

  1. Gravatar of Mishima Mishima
    14. July 2009 at 15:41

    Yes, Houellebecq is French. He’s also a pornographer masquerading as a deep thinker.

  2. Gravatar of TGGP TGGP
    14. July 2009 at 16:45

    Mencius Moldbug defines money as a bubble that never pops. Making your proposal right up the Fed’s alley.

  3. Gravatar of mark mark
    14. July 2009 at 16:53

    Good idea. If I were a commercial bank or perhaps an insurance company I would sell out of the money puts on the market up to 100x my capital base since the Fed would “have my back.” I’d get fabulously wealthy underpricing risk.

  4. Gravatar of StatsGuy StatsGuy
    14. July 2009 at 17:12

    It’s important to separate two distinct problems with the US economy today:

    1) An utterly moronic macroeconomic situation that has 9.5% of people unemployed, another 1.5% having given up, and 5-6% more underemployed, even while capacity utilization is low and (much like the Grapes of Wrath) we have productive assets sitting unused.

    2) Everything else that is wrong, which NGDP targeting does not necessarily fix. The moral hazard problem with banks? NGDP targeting doesn’t address that. The energy problem? NGDP doesn’t target that. The health care debacle? Nope. The failure of govt. to invest in externality-producing basic technology that drives _real_ economic growth over a 20 year timeline. Er, nope. (Growing NGDP doesn’t necessarily mean life is getting better; just that we don’t have unnecessary liquidationism, deflation, and capacity underutilization.) The foolish misuse of our armed forces? Sadly, nope. The corruption of technical scientific advice in govt. agencies? Don’t make me laugh.

    These other problems will still cause great grief. NGDP might grow at 5%, but life can still get worse.

  5. Gravatar of StatsGuy StatsGuy
    14. July 2009 at 17:21

    Re ssummer’s comment (previous post) that: We know that when Democrats win the presidency, the stock market drops 2%…

    We also know, based on 100 years of data, that when Republicans presidents run the economy, we run larger fiscal deficits AND experience slower real GDP growth.

    Obviously, I’ve discovered a flaw in the EMH…

    Interestingly, we also have strong evidence that the economy does significantly better under Presidents who cheat on their wives than Presidents who are faithful.

    http://thenoosewire.blogspot.com/2007/03/study-shows-men-who-cheat-on-wives-make.html

  6. Gravatar of David Pearson David Pearson
    14. July 2009 at 18:21

    If you abolish the business cycle, what will be the level of leverage you would expect in the economy? That’s right, it will be really, really high. Higher even than the 350% debt to GDP ratio we managed to reach in 2008.

    So follow your logic. If the market never expects NGDP to fall below 5%, then leverage piles up, and up, and up. 500% debt to GDP? 600%?

    No, you’ll argue. At some point, pile up enough debt and the system becomes unstable. It self-corrects through defaults and bankruptcies. So we’ll never get to 500%.

    Exactly. This “self-correction” mechanism is called the credit cycle. I will not be cured by having 5% NGDP carved on every market player’s forehead. That’s because with sufficient leverage, the system gives way to defaults even with 5% NGDP growth, in much the way that we saw subprime defaults escalate even with 5%+ HPA in late 2005.

    So, tell me, what’s your prediction for debt to GDP under your “permanent bubble” regime, and if it doesn’t continue, exponentially, then what is the corrective mechanism?

    By the way, the process under which debt to GDP continues to expand in a boom/bubble is called Ponzi finance. You may have heard of it before…

  7. Gravatar of Mishima Mishima
    14. July 2009 at 18:33

    TGGP, for someone who dislikes and critiques both Austrian economics and Mencius Moldbug’s ideas, you sure do seem to link to his articles on economics a lot.

    What gives?

  8. Gravatar of Rob Rob
    14. July 2009 at 18:35

    StatsGuy:

    “when Democrats win the presidency, the stock market drops 2%” would not reject the EMH, just as the Super Bowl Indicator would not (although the January Effect would), because the outcome is not known in advance and therefore cannot be discounted in advance.

    But statistical inference falls apart when you have already peeked at the data, which is what the Election/Stock Market correlation does. I realize that economists usually have the luxury of lacking data points, and therefore don’t normally have to prove their hypotheses. But it’s a death wish for a trader not to prove a hypothesis with some degree of confidence. Economists get to be hedgehogs while traders have to be foxes. But if economists want to convince traders that they are being honest with their data, they need to demonstrate they can think like one every now and again.

    Sorry, I’ve had too much whiskey. The piano is drunk, not me.

  9. Gravatar of Lord Lord
    14. July 2009 at 18:59

    Monetary policy may only be able to target one thing at a time but regulatory policy could target another. I would lay the trigger for the 2001 recession as dashed tech expectations even though the Fed was too slow to react to them. I have doubts the Fed could produce a permanently high stock bubble unless it was willing to put a floor under stocks by buying them, though I wouldn’t object to its trying. I don’t think they should have bought pets.com because it was failing though. Personally, the only bubbles that concern me are credit/debt bubbles since they destroy wealth through consumption leaving debt in its wake, whereas equity bubbles merely redistribute it.

    I wonder what you think of those that predicted this crisis on a flow of funds basis such as at nakedcapitalism, High time for a coherent macroeconomics. I realize housing was being digested but as housing was the economy this decade, there wasn’t anything to take its place.

  10. Gravatar of Jon Jon
    14. July 2009 at 19:22

    2. Monetary policy can only focus on one target at a time.

    Like the price if Milk or Gold :)? Obviously we can build objective functions. So you’re saying that the ‘Fed’ cannot nail more than one thing at a time. Doesn’t that imply that the Fed has only one lever?

    I disagree. The Fed can nail any basket of asset prices just as any other speculator would provided they intend to make that asset more scarce relative to money–and if the gradually accumulate a position (like their gold reserves…) they can do the opposite for a time.

    Can you clarify what you mean here?

    They don’t occur randomly, but rather tend to occur when the overall macro economy is doing well; when there is steady growth and low inflation. The late 1920s, the mid 1960s, the late 1980s, the late 1990s, and 2005-07. The two biggest stock booms (1928-29 and 1999-2000) correspond to what are arguably the two best periods in American macro history.

    I was under the impression that occurred during periods of rising inflation expectations. Please substantiate your remarks…

  11. Gravatar of Current Current
    15. July 2009 at 01:46

    You were right, that was an unspeakable horror.

    Come on, you can’t be serious about this. I really don’t know where to start, I could use the economic ideas of Hayek and Mises to show why this is a bad idea. Alternatively, though I could use the ideas of Fisher, or Joan Robinson, or Adam Smith, or Marx or my mate Alan the stoner.

    Think about it, what role do expectations play in an economy?

  12. Gravatar of Bill Woolsey Bill Woolsey
    15. July 2009 at 01:52

    Pearson:

    It seems sensible to me that if macroeconomic policy were better, then optimal capital ratios would be lower for all sorts of businesses. Where there are plenty of sourcs of flctuations for an individual firm, there would be one less, and so the capital cushion needed to avoid the cost of bankruptcy would be less.

    You have given no reason to believe that optimal capital ratio would fall and fall until it gets “too low” and then rise. That is, the notion that steady nominal income growth would lead to a credit cycle of any sort is wrong.

    Also, I think you need to think about what changing capital ratios mean. Do you understand? What exactly happens when the debt to income ratio gets high and stops growing?

    Are you claiming that nominal income must drop? Are you claiming that real output and income must drop? Are you claiming there must be more real consumption and less real investment? (More production of consumer goods and less production of capital goods?) Are you claiming that wealthy households must consume more and poorer households must consume less?

    When leverage stops rising or starts to fall, what do those who don’t lend do with their money? What do those receiving loan repayments do with their money?

    When we say that firms are overleveraged, we are not saying that they have all expanded too much. We are saying that they should have financed their expansion with equity. We are saying that investors should hold stocks rather than bonds.

    If households are only willing to finance real investment by holding bonds, then what are they supposed to do? Well, the answer is consume. What else is there?

    On, or prehaps they could choose to enjoy more leisure.

    The “theory” that people finance spending by borrowing, so we need lots of borrowing to fund ever growing spending, and when debt gets too high, there can’t be more borrowing, and so their can’t be more spending, so spending needs to fall…

    It is wrong. It is ignoring the fundamental relality that there is a lender for every borrowing.

    Nearly all spending is financed output out of current income. There is no reason why all of it couldn’t be. That is, if we can produce it, we will earn enough income to buy it, and there is no need for debt to finance it.

    Debt shifts expendenditure between and amount households and firms. It allocations the composition of demand and the allocation of resourcs.

    Now, if we have a “recession,” because people have accumulated all the wealth they want and now they want to retire either fully or partly, how is that a problem? Sure, less labor input, less output, less income.. yes, but so what?

    If people what to consume now, and not buy capital goods now, if this is so extreme that the capital stock falls, then output will fall in the future. More consumption now, less investmnet, less production and consumption in the future. Why is this a problem?

    No. I beleive that you are implicitly assuming that repaid debts result in an increase in the demand to hold money. If lenders don’t lend the money, they hold it. And so, you ball up a credit cycle with monetary disequilibrium.

    If index futures convertibility works, then the quantity of money will adjust enough so that it will accomodate any change in the demand for money. If there is a credit cycle, can only impact the composition of demand and the allocation of resources in a way that responds to the preferences of consumers.

  13. Gravatar of Adam P Adam P
    15. July 2009 at 02:30

    Current: “Think about it, what role do expectations play in an economy?”

    I don’t get it, are you being sarcastic or daft with that statement?

  14. Gravatar of ssumner ssumner
    15. July 2009 at 04:22

    Mishima, The term ‘pornographer’ is used for both “offensive” and “sexually explicit” (which are two very different concepts.) I’m not sure what you mean here, but I suppose both terms could apply to Houellebecq. “Deep thinker” is also a fuzzy term. I like interesting thinkers, and pragmatic thinkers. A lot of “deep thinkers” are neither. He doesn’t strike me as a great novelist, but it seems to me that he has a few interesting ideas, which is more than you can say for most other writers.

    TGGP, Am I right that Moldbug doesn’t mean it in a good way? :)

    Mark, Yes, making the economy more stable will make banks take more risks. An economist once argued that every car should have a dagger mounted on the steering wheel, pointing directly at the driver. He argued that it would make people drive more safely, and fewer pedestrians would be killed. What do you think of that idea? Seriously, we do need to reduce moral hazard in banking, I agree that there is a bias toward too much risk. My proposal might encourage more risk taking.

    Statsguy, I totally agree.

    Note that in May 2008 the subprime crisis was fully priced into stocks, but the market was still near its all time high (down about 10% from the peak.) So while “real factors” do affect stock prices, it is nominal shocks that do the greatest damage. So I still think that average P/Es would be much higher with NGDP targeting, although all those other problems would continue, as you say.

    Statsguy, What you say is correct, but only the statistic I quote is relevant. The actual performance is highly erratic, and I don’t find the numbers to be statistically significant. It’s just another data-mined anomaly.

    David, The mistake you make is to assume that financial crises make the economy unstable. They do not. It is NGDP fluctuations that make it unstable. So let’s suppose everyone gets leveraged, and we have a financial crash. Big deal. Under NGDP targeting it merely redistributes a little wealth, the economy keeps chugging ahead. I find that it is very difficult to get people to shake the misconception that our current recession and high unemployment is caused by the financial crisis. It isn’t, it’s caused by falling NGDP. In fact, much (not all) of the financial crisis itself is caused by falling NGDP, and would never have happened with NGDP futures targeting.

    Rob, I think I agree. BTW, the 2% figure comes from comparing betting odds changes to stock prices. The evidence is really strong.

    Lord, Until everyone in America is as rich as Bill Gates, there is always “something to take its place.” I’d like a bigger consumption bundle, who wouldn’t? So we should never have a recession due to too little AD.

    By high plateau, I do not mean stable stock prices. They would still fluctuate, but the fluctuations would be around a higher trend line.
    Jon, Yes they can nail any basket of assets (like the CPI) but I consider that “one thing.” They cannot control each individual price within that basket.

    Jon, I believe that inflation expectations were relatively low and stable during those periods. The inflation rates (and expectations) ranged from zero in the 1920s to a high of about 5% in the late 1980s. But even that was regarded as low compared to the 1970s, so stocks did well.

    Current, Before you revoke my Austrian economics membership you better check out Hayek’s views. He and I both favor NGDP targeting, so I don’t think my idea is quite as anti-Austrian as it sounds. Although I will admit that the tone is very un-Austrian.

    Thanks Bill, I agree.

    Adam P, Yes, I didn’t get that either. My ideas may be crazy, but I don’t see how anyone could claim I ignore expectations. The opposite charge is often made, that I put too much faith in targeting expectations. BTW Adam, I am scheduled for a webinar discussion/debate with John Cochrane today. I will let you know the address of the website where it will be stored when it is up and running. Should be interesting.

  15. Gravatar of David Pearson David Pearson
    15. July 2009 at 05:37

    Scott,

    I think you’re saying that financial crises can occur in the absence of falling NGDP:

    “So let’s suppose everyone gets leveraged, and we have a financial crash. Big deal.”

    So its not that you think financial crises are entirely preventable; just that with NGDP targeting they are relatively benign — just distributive.

    The problem is with what “distributive” means to an economy. How much friction does the “distribution” process cause, and what is the impact on animal spirits? Very little, you would argue, because animal spirits will be cheered by rising NGDP.

    There is another view. Each successive boom results in more mal-investment (Bill Woolsey) as investors chase exceedingly low unlevered expected returns (which they inflate using leverage). The mal-investment leads to loss expectations which dampen animal spirits and reverse the leverage gains. Each time the Fed tries to drive up NGDP expectations in this environment, it will get more “N” and less “GDP”, if you will. That’s because without reducing asset prices, unlevered REAL returns do not increase enough to promote higher investment.

    Anyone familiar with emerging markets knows that the real impact of repeated monetary stimulus is that it makes it difficult to forecast REAL returns from long-term investments. Yet your NGDP targeting regime assumes that nominal returns lead to real returns. This is an exceedingly optimistic assumption.

  16. Gravatar of Bill Woolsey Bill Woolsey
    15. July 2009 at 05:39

    Scott:

    It isn’t a bubble. By definition… (lol.)

    But really, that stocks on average would be worth more because of less volatility of earnings, because of stable growth in nominal income… has nothing to do with a bubble.

    Mark:

    With index futures targetting on nominal income, the Fed would not “have your back” regarding investments in the stock market.

    The out of the money puts on stock prices would only “work” if stock prices can never fall as long as nominal income grows 5% (or 3%, or whatever.)

    If stable growth in nominal income really did lead to a perfectly predictable stock market, then why would anyone buy puts?

    Sumner is really just saying that he would never favor a monetary policy that would lower expected nominal income in order to push down stock prices.

    Monetary policy should always keep nominal income growing at a steady pace (he likes 5%,) and if that means stock prices are really high, well, fine.

    But, the other side of the coin, is that if stock prices fall, then monetary policy should do nothing about it. It should just keep nominal income growing 5% (or 3%) a year.

  17. Gravatar of David Pearson David Pearson
    15. July 2009 at 05:50

    Scott,

    BTW, as always, I am rooting for you to be listened to. I think the Fed will ultimately adopt your view in order to fight what it views as dangerous deflation. In doing so, it will create the threat of dangerous inflation, and I am positioned for that as an investor.

    So you would buy stocks anticipating higher real returns, I buy vehicles that benefit from higher nominal returns. Are you that optimistic? If you’re worried the Fed won’t listen to you (unlikely), then buy 30-yr Treasuries (levered) as a hedge against deflation, or merely short gold.

    I have a short stock/long gold portfolio, you would do the reverse. Have you? I wonder why you wouldn’t…

  18. Gravatar of Thruth Thruth
    15. July 2009 at 07:06

    “Note that in May 2008 the subprime crisis was fully priced into stocks, but the market was still near its all time high (down about 10% from the peak.) So while “real factors” do affect stock prices, it is nominal shocks that do the greatest damage. So I still think that average P/Es would be much higher with NGDP targeting, although all those other problems would continue, as you say.”

    Scott: Even though we agree that something pivotal happened in mid-late 2008, I still think you dismiss the role of the financial crisis too easily. Sure, you can make the case that if we had NGDP futures targeting (or a Fed that basically acted as if it did) the crisis could never spill over. But we don’t live in that world. Look at all the indicators of risk: VIX, Ted spreads etc. They start ramping up in mid 2007. Clearly investors perceived the risk of economy wide calamity, Fed driven or otherwise, well before the calamity actually happened. So in the sense that we still live in the second best world, the credit cycle matters.

    “David, The mistake you make is to assume that financial crises make the economy unstable. They do not. It is NGDP fluctuations that make it unstable. So let’s suppose everyone gets leveraged, and we have a financial crash. Big deal. Under NGDP targeting it merely redistributes a little wealth, the economy keeps chugging ahead. I find that it is very difficult to get people to shake the misconception that our current recession and high unemployment is caused by the financial crisis. It isn’t, it’s caused by falling NGDP. In fact, much (not all) of the financial crisis itself is caused by falling NGDP, and would never have happened with NGDP futures targeting.”

    It strikes me that this “redistribution” implicit in your NGDP futures proposal is in some sense an attempt to correct a market failure: the apparent unwillingness of investors to take equity positions. (Think about the literature on the equity premium puzzle — why is the premium for holding equity over bonds so large?). By doing so we create a more stable economy, which should lead to lower risk premiums (as embodied by P/E ratios etc).

    Like David, I do wonder if moral hazard isn’t being swept under the rug. But perhaps bond holders will take care of that in the way they price debt in an NGDP futures world. I they will recognize the implicit redistribution that takes place when the system takes on too much leverage, and price debt accordingly.

    The proposal has other interesting impact on holders of non-govt debt. Because it rules out deflation, there are no “real” windfalls due to deflation. But eliminating deflation reduces the risk of system-wide default, which debtholders are incredibly averse to.

  19. Gravatar of Current Current
    15. July 2009 at 07:30

    Current: “Think about it, what role do expectations play in an economy?”

    Adam P: “I don’t get it, are you being sarcastic or daft with that statement?”

    Scott: “Current, Before you revoke my Austrian economics membership you better check out Hayek’s views. He and I both favor NGDP targeting, so I don’t think my idea is quite as anti-Austrian as it sounds. Although I will admit that the tone is very un-Austrian.”

    * I agree entirely that the Fed can’t prick Stock Market bubbles without causing much other mayhem.

    * Nobody denys that high stock market valuations are associated with good conditions. However, high valuations due to good conditions are not bubbles, as Bill notes. They are connected with proper valuation of companies, which is exactly what a bubble isn’t.

    * You say “I define good times as ‘5% expected NGDP growth.'” What has that got to do with “good times”?

    * You say “As long as stable nominal growth with low inflation is expected, then stock prices should be able to stay at relatively high levels””Irving Fisher’s famous ‘permanently high plateau.'” Why? NGDP doesn’t necessarily relate to RGDP, which doesn’t necessarily relate to good stock price valuations. I think you are unconsciously falling back on the Keynesian idea that it is imperative to scare money holders out of holding money into holding stock and bonds. As I have pointed out before although there is a social cost in individuals holding money there is also a social benefit from it.

  20. Gravatar of TGGP TGGP
    15. July 2009 at 07:45

    Mishima, perhaps due to their rejection of mainstream academia and all its works, the Austrians put a lot of effort into making their ideas accessible to laymen on the internet. Also, I find them interesting even if they are mistaken.

    Scott, he doesn’t like the Fed’s involvement, but he’s clearly comfortable with the idea of a never-ending bubble, and in fact thinks it perfectly natural in the case of money.

    Off-topic, but Scott Fulwiler has a rejoinder on negative nominal interest rates here.

  21. Gravatar of Current Current
    15. July 2009 at 07:58

    This business of “redistribution”.

    Certainly stock market falls redistribute wealth. More importantly though they are associated with changing expectations of future income.

  22. Gravatar of Rob Rob
    15. July 2009 at 08:29

    What a boring world this would be if we kept the mother of all bubbles going!

    The only stock investors to benefit would be those who got into the market before the permanent multiple expansion. After that, with little equity premium, returns would suck.

    Maybe Vic Neiderhoffer is right and stock market drops exist so the strong hands can shake the money out of the weak ones. Bernanke is in the pocket of Goldman and so handed them this great buying opportunity! If your idea of targeting NGDP is suddenly adopted — the conspiracy theory is complete!

  23. Gravatar of Current Current
    15. July 2009 at 08:52

    Rob: “What a boring world this would be if we kept the mother of all bubbles going!”

    I wouldn’t lose sleep over the possibility of it happening anytime soon.

  24. Gravatar of Bill Woolsey Bill Woolsey
    15. July 2009 at 09:02

    I have trouble with Fulwiler’s blog as well. I get errors when I open it. And I couldn’t get my comment to post.

    Did I miss something? Does he understand that as checkable desposits expand, the amount of reserves that banks must hold rise?

    I was just flabergased by his notion that the “tax” on currency would only cause the increased use of debt cards. The whole point is to get negative nominal yields on short, lower risk assets. That would include the FDIC insured deposit.

  25. Gravatar of Mishima Mishima
    15. July 2009 at 09:42

    TGGP,

    I ask because while I know that you disagree with Mencius and the Austrians, I’ve never seen you critique their ideas on econ on your own blog or on others, aside from a few comments here and there explicitly stating that you disagree with them.

    It seems that more often you will link to Mencius’ Austrian views in a tacit endorsement of such views, and for someone who purportedly disagrees with such views I found it somewhat strange.

    I only recently started reading the Austrians, and in my quest to be ever vigilant against confirmation bias I am always seeking contrary arguments.

    Please do point me to any arguments you’ve made against the Austrians or Mencius specifically. I’d be glad to read them.

  26. Gravatar of StatsGuy StatsGuy
    15. July 2009 at 09:59

    My only objection to negative nominal rates is the issue of the closed vs. open economy. So I pose this question to others who understand the issues better than I:

    I now get that base money is always held in either vault cash or on deposit with the Fed… in a closed domestic economy. Is this still true of an open economy?

    In other words, if we have negative rates on reserves, and banks move assets to reserves with non-domestic banks (at 0% interest rate), what happens?

    It seems to me as if the foreign economy that is receiving the reserves gets the benefit of an expanding money supply EVEN WHILE the local currency gets strengthened… I think this is why the Fed fears capital flight. It could trigger simultaneous depreciation of the dollar, appreciation of foreign-commodities (e.g. oil), AND shrinking of the domestic money supply. (I’m certain there’s something wrong with this logic – but what?)

  27. Gravatar of Bill Woolsey Bill Woolsey
    15. July 2009 at 10:15

    Pearson:

    I have no idea how stable growth of nominal income is supposed cause ever worsening crises and negatively impact animal spirits.

    I don’t have any idea why changes in the ratio of debt to GDP are a problem.

    Personal saving in the U.S. has been positive nearly every quarter for the last decade. But consumer debt has risen. How is that possible. Well, some households are lending to other households. And, if they stop, then the households that were borrowing must now consume less, no more than their incomes and the households that were lending must now consume more. Household debt can stop rising, but saving and consumption remains the same. How is this supposed to involve malinvestments and hurt animal spirits?

    Can household debt fall? Yes. Households that were consuming more than their income now save and pay down debts. Households that have been accumulating assets do less of it. The don’t actually need to dissave, then just save less. (Remember, saving has been positive all along.)

    Household debt falls, consumption stays the same. Saving stays the same.

    What about firms? Can they deleverage and still invest? Sure. Some firms own bonds issued by other firms. The firms that sold bonds pay them off. The firms that bought the bonds use the funds to buy capital goods. Or, of course, they could be stock in other firms.

    Similarly, if firms that sold bonds pay off housholds, the households can use the money to buy stocks.

    It is possible to delever all you want without there being any change in the allocation of resources or “malinvestment.”

    But, that doesn’t mean there won’t be a change in the allocation of resources. Maybe there is. But I don’t see this as malinvestment.

    If people decide they like coffee more than ice cream, then there is going to be a reallocation from tastee freeze to starbucks. If they go back the other way, then the reallocation goes the other way. It is true that some specific capital goods may be wasted. There are adjustment costs. But that is the way the world is.

    This “leverage” and “excessive debt” business is all based on the assumption that money not lent is held. Money that is repaid is held. The quantity of money is “fixed” and so, when more money is held, there is an excess demand for money. And aggregate demand falls.

    Monetary disquilibrium is always hidden in this stories.

    When you take out the monetary disquilibrium you have reallocation stories. And reallocation happens all the time. But when there is no monetary disequilibrium it isn’t that much of a problem.

    For example, it is true that the growth of Walmart may have turn many mainstreet merchants. Superwalmarts have hurt unionized grocery stores. Yes, it is troubling. Do we worry about all the malinvestment in stand alone grocery stores or downtown districts? I don’t. Do we worry about animal spirits being dashed. I mean, look at all the small businessmen who fail because of the competition?

    Or is the “distribution” we are talking about the ability of people to live on investment income? Getting nominal income to target may imply very low interrest rates on short term low risk securities. As I aways say, if you want return, you have to take risk.

  28. Gravatar of Bill Woolsey Bill Woolsey
    15. July 2009 at 10:36

    Statsguy:

    You missed out on the part of base money used as currency.

    I am not sure what you mean by banks moving assets to reserves with nondomestic banks with zero percent reserves. If you mean, holding deposits with foreign banks, then… well yes. they could. But I am afraid that you seem to think that what is happening is that the U.S. bank is withdrawing its gold from the Fed (which is charging this pentalty rate) and then moving it to the Bank of England. Well.. there is no gold. When the U.S. bank purchase some foreign security, including foreign bank deposits) some other bank ends up with the balance at the Fed, still with the penality rate.

    Anyway, the dollar would fall in value. So you are right about that part. And what happens is that U.S. exports end up with more balances in their checkable deposits, which increase the amount of required reserves.

    Anyway, the lower value of the dollar increases the demand for U.S. exports. Yes, it does raise the prices of foreign goods, which increases the demand for import competing goods in the U.S.

    This raises the demand for U.S. goods. Presumably it results in more output for them as well as higher prices. And so, the lower value of the dollar is a means by which the penalty rate on excess reserves raises aggregate demand.

    Nominal income in the U.S. is supposed to be measured using U.S. output and the prices of U.S. output, not the prices of the imported goods. But it is probably true that the inflation rate in terms of what people will buy will be higher than the inflation rate for U.S. goods.

    Anyway, we need to keep up interest rates so the dollar won’t fall, so import prices won’t rise is not consistent with keeping nominal income in the U.S. growing on target.

    If we keep nominal income growing on target, then the value of the dollar will change leading to changes in import prices and so real incomes through the U.S. terms of trade. It is something we have to live with.

    Hmm.. Making sure that people with portfolios of safe, low risk aseets can continue to buy rolls royces, Russian caviar, and french wine.. inconsistent with nominal income targetting.

  29. Gravatar of StatsGuy StatsGuy
    15. July 2009 at 10:59

    Bill, thank you –

    I agree that this would raise revenue for US exporters immediately, but it would raise costs for commodity related US imports (and we import more than we export). This is triply so if the cost of commodities rises due to anticipated US inflation but the cost of US exports lags this increase (or does not increase, since many of our exports are service or knowledge goods).

    Also, the US increasing exports of goods and services will take time (the J curve), so in all likelihood our current account deficit would increase in the short term before falling in the long term. This would reinforce the exchange pattern.

    I suspect if the Fed was targeting NGDP using those tools, and a dollar run occurred, it would eventually stabilize. Ultimately, other central banks would intervene, commodities would hit a ceiling (demand would drop), and we fortunately don’t owe our national debt in foreign currency. The dollar would overshoot, but this might be necessary to start a path to a sustainable recovery…

    I have to wonder, then – perhaps the reason the Fed tightened in August 08 was for fear of a currency run, that they perceived manifesting through the oil price peak? Fed signaled tighter money to combat the dollar run/oil bubble, which popped rapidly…

    Personally, I’d rather be paying $4.20 a gallon of gas and have 6% unemployment and a 401k I’m not scared to look at. Besides, $4.00 gas was actually giving the country an incentive to invest in a real energy plan.

    But, here’s another question – what happens if US banks directly buy and hold foreign govt. bonds with the easy money from the Fed?

  30. Gravatar of StatsGuy StatsGuy
    15. July 2009 at 11:07

    Also, I suppose there’s also the issue of banks buying foreign assets with physical dollars that then stay abroad… For example, with dollarized economies… If demand for those physical dollars does not exogenously increase, then those dollars leave the domestic economy while taking down the dollar.

    Personally, my reaction is: Good, let’s devalue the dollar and turn the US into an export-led economy like Germany and China. Ahoy mercantilism! That puts me in esteemed company… Alexander Hamilton, Abraham Lincoln…

    Men who died by gunshot.

  31. Gravatar of Jeremy Goodridge Jeremy Goodridge
    15. July 2009 at 13:06

    Scott:

    Please comment on Krugman’s latest post on deficit spending. His claim is that the key difference in policy between now and the Great Depression is deficit spending. Do you agree? While it’s true the US deficit has risen more this time around (although he presents ZERO stats on this — which is another annoying thing about the piece), Fed stimulus is also much higher than it was during the Great Depression. While you (and I) may not think the Fed is doing enough, there’s no question it’s doing a lot more than it did in the early 30’s. So, is that effect just irrelevant — does he really believe that? Would we be in the same position now if the Fed had behaved as it did during the Great Depression but the Treasury had spent just like they are now? I don’t know how that can be — almost all of the Treasury’s spending has come in the form of transfer payments NOT actual govt consumption. So, is his argument that the transfer payments from the Fed have a much lower effect than the ones from the Treasury? If so, he certainly didn’t say that.

    And finally, the point that in 1933, we started turning around NOT because of any fiscal stimulus but because of massive monetary stimulus is just totally ignored by him. That suggests the multipliers for monetary policy are, in fact, very large.

    Jeremy

  32. Gravatar of TGGP TGGP
    15. July 2009 at 17:09

    Mishima, we may be getting off topic. In which case I apologize to the others here. Anyway, my biggest problem with the Austrians is epistemic. LessWrong has a post on “Awful Austrians” that sums it up. You could also check out Bryan Caplan’s “Why I Am Not An Austrian”.

    I don’t comment as much at MM’s blog as of late. When I did I gave a lot of criticism though. You can also find some running arguments between us in the archives of my blog. Getting back to epistemics, you’ll note that he has the same susceptibility to kooky conspiracy theories (Obama didn’t attend Columbia or write Dreams) as many in the Rockwellian crowd (whom MM would consider to be the only real Austrians).

  33. Gravatar of Lord Lord
    15. July 2009 at 18:37

    Bill:
    As for the hazards of debt, I offer http://www.levy.org/vdoc.aspx?docid=866
    This was 2006 so housing was still being worked on but there was little doubt recession lay ahead. Even if the Fed targeted ngdp, I think both the Fed and the economy would have had difficulty responding to the impending discontinuity rapidly enough, not that they shouldn’t have tried and not that it wouldn’t have been lessened if they did. Sometimes the changes may simply be too wrenching to avoid a recession.

    The other problem is the debt was too much to ever have been repaid, even under constant ngdp growth as it was based not on income but spiraling asset prices, so unless the Fed started targeting housing prices, it would still have a problem.

  34. Gravatar of Jon Jon
    15. July 2009 at 23:05

    We know that if the price-level rises above trend that its deleterious for Fed to revert to the original trend-line–whether that trend is zero-inflation or 2% or some other number. This is of course nothing more than the old Austrian argument.

    I think this is very uncontroversial. Modern policy is ‘memoryless’.

    Now what about the NGDP policy. If NGDP drifts off-trend, do you expect the Fed to be memoryless? I don’t think its necessary in the same way…

  35. Gravatar of Current Current
    16. July 2009 at 03:13

    On his blog Robert Murphy has made the interesting claim that the CPI figures are been fudged…

    http://consultingbyrpm.com/blog/2009/07/another-month-another-big-price-spike.html

    I don’t know enough about the US to know if this is true. I know that I certainly don’t believe the UK or Irish government’s CPI figures.

  36. Gravatar of Bill Woolsey Bill Woolsey
    16. July 2009 at 03:47

    Jon:

    Sumner (and I) advocate level targeting for nominal GDP–a growth path for nominal GDP. The expected level is targeted in the future.

    For Sumner, if NGDP somehow jumped up 6% above target this quarter, then it does imply that the target for NGDP one year from now would be 1% below the current level.

    I favor a 3% path, so it would only have to be more than 3% above target right now to require that the target for next year to be less than its current value. So, if it somehow was 4% above target, then it would need to actually fall by 1% during the next year.

    Of course, more to the point, for quarter 1 2009 was less than it was in quarter 1 2008. Sumner thinks it should be 5% (and I am already settling for 3% higher.) So, yes.. I favor getting it back to its passed growth path rather than having it grow at the target rate from its current depressed level.

  37. Gravatar of Jim Jim
    16. July 2009 at 05:54

    Off topic

    The Cleveland Fed has a new article out on inflation expectations here http://www.clevelandfed.org/Research/trends/2009/0809/01monpol.cfm

    Would you mind commenting, especially on how they adjust for “liquidity concerns”?

  38. Gravatar of ssumner ssumner
    16. July 2009 at 09:12

    I am way behind because my home computer broke. I only have about 20 minutes to catch up on comments and show that I am still alive, and then I need to go out to shop for a new computer.

    David, We saw a financial panic without sharply falling NGDP. It occurred in late 2007 and early 2008, and it was no big deal compared to what happened later. (Obviously it was very big in an absolute sense–maybe close to a trillion in losses, but small relative to the more recent crash.) Unemployment rose modestly.

    I don’t think the experience of developing countries is at all relevant to my proposal. 5% NGDP targeting is not a radical shift from what we have been doing in the US. Indeed according to Austrian theory NGDP targeting is actually superior to inflation targeting. So even if it is not perfect, it would be an improvment over current practice, which features highly erratic and unpredictable NGDP fluctuations.

    I certainly don’t think doing this it would cause malinvestment. It wouldn’t completely prevent it, but then no other proposal would either. I still think many Austrians don’t understand the superneutrality of money. Permanent changes in the trend rate of inflation do not affect the business cycle, or the amount of malinvestment. The Fisher effect neutralizes inflation that is anticipated.

    Bill, of course there is something slightly absurd about the debate over bubbles:

    1. A bubble could be defined as a rise and fall in an asset price. But that’s totally stupid, asset prices are supposed to rise and fall. But even so, many use this defintion.

    2. Or it could be defined as an asset price that becomes too high according to some arbitray criteria. The more intelligent bubble theorists (Shiller, etc) use this definition. They say a bubble is defined as a P/E ratio above X, or something similar, which they derive through historical study. I think this is also silly, but at least it is much less silly. Shiller’s model does work pretty well looking backward. So I am saying “OK, if that’s your definition, then I think we can create a permenant bubble.”

    Maybe this is obvious, and your answer was tongue in cheek, but I thought I should spell it out in case anyone didn’t see what I was getting at in the admittedly absurd title of the post.

    David, Remember that I believe in the EMH, so why would I try to beat the market? My only gamble has been Asia, I have always been heavily invested in Asian (especially Chinese) stocks, as I like the potential of Asia. It’s been a rollercoaster ride.

    Thruth, I don’t quite follow your argument. Yes those TED spreads increased during late 2007, but doesn’t the fact that the stock market didn’t collapse tell us that investors thought the Fed was capable of preventing the damage from spilling over into the broader economy? Regarding the far more severe financial crisis of late 2008, I totally agree that it is far-fetched to believe that any Fed policy could have totally insulated the economy from a crisis that severe. But remember that my argument is that the second and much more severe crisis would never had occurred with NGDP targeting. Indeed the reason that stock market didn’t collapse after Lehman is that it still had hope that the Fed would act as aggressively as it had in January 2008, when it cut rates 125 basis points in less than 10 days. The Fed didn’t do this. Only then did markets crash.

    We know that the Fed can always make NGDP go up with an expansionary enough monetary policy. If there is still a recession it is due to a sharp fall in AS. But I’ve never seen a US financial crisis that primarily worked through AS. I’ve never seen a financial crisis make RGDP fall sharply, without also depressing NGDP.

    How long it takes me to catch up with the others, depends on how long it takes me to buy a new computer and set it up.

    Thanks for your patience.

  39. Gravatar of Thruth Thruth
    16. July 2009 at 09:58

    “but doesn’t the fact that the stock market didn’t collapse tell us that investors thought the Fed was capable of preventing the damage from spilling over into the broader economy?”

    or alternatively, that there were equally large risks that the Fed/Govt would do to little (deflation) or too much (inflationary windfall to stockholders), hence increased stock mkt volatility (as seen in the VIX).

    Mind you, the stock market did gradually drift down from mid2007, suggesting that the Fed doing too little was being seen as an increasingly likely possibility.

  40. Gravatar of 123 123
    16. July 2009 at 13:27

    “So I am saying “OK, if that’s your definition, then I think we can create a permenant bubble.” ”

    You can’t create a permanent bubble, you just can reduce the frequency of anti-bubbles.

    ” My only gamble has been Asia, I have always been heavily invested in Asian (especially Chinese) stocks, as I like the potential of Asia. ”

    So what about Chinese monetary policy. Is chinese NGDP stable enough to create a permanently high plateau?

  41. Gravatar of StatsGuy StatsGuy
    16. July 2009 at 14:19

    THIS is why the Fed/Treasury are working so hard to show that they are good little Hooverites:

    http://www.bloomberg.com/apps/news?pid=20601087&sid=axGuZ_0J4AsA

  42. Gravatar of rob rob
    16. July 2009 at 16:18

    “The two biggest stock booms (1928-29 and 1999-2000) correspond to what are arguably the two best periods in American macro history.”

    Why were they also inarguably the worst times to invest? Why need the best of times stand so starkly against the very worst? (Unless you’re a Taoist?)

    You are not allowed to use the word bubble in your answer.

  43. Gravatar of ssumner ssumner
    16. July 2009 at 17:41

    Current, I’m not even sure bubbles exist. It’s other people that suggest high stock prices are bubbles. You say it’s not a bubble if the high prices are justified. According to whom? You, or the consensus of thousands of very sophisticated Wall Street investors? Other people say 1929 was a bubble, but investors didn’t think so, and later researchers found that those prices would have been very justified if we hadn’t gone into a depression.

    I admitted in the post that other things can go wrong (supply shocks), but I also said most of the spectacular crashes were due to demand shocks. So I still think 5% NGDP targeting would lead to a permanently higher trend line (plateau may be the wrong word. Surely Fisher didn’t mean prices would never fall, but he meant they’d fluctuate around a higher trend than before. BTW, Shiller was wrong in 1996, stocks had reached a permanently higher plateau (so far, knock on wood.)

    TGGP, Regarding Fulwiler, his reply section never works. I stopped reading after these two assertions:

    1. He claimed the “monetarist excess cash balance mechanism” was the money multiplier. That will be news to Keynesians who teach the money multiplier.

    2. He claimed that only the Fed can affect the total amount of bank reserves. Actually I got a roll of quarters today from the bank, and the instant I did so aggregate bank reserves fell by $10.

    Current#2, Yes, but stock market falls don’t cause changing expectations of future income, they reflect changing expectations.

    Rob, Within a week I plan another post demolishing the view that “strong hands” can predict the stock market.

    Bill, Yes, there are lots of problems with Fulwiler’s post.

    Statsguy. This was discussed earlier, and all I recall is that it isn’t a problem. The tax would apply to US dollar bank reserves, wherever they are held. Of course cash would be exempt.

    Jeremy, Thanks to you and Dilip I did comment on Krugman. I’ve already talked a lot about the Depression in other posts, so I focused on slightly different issues here.

    Lord, Yes, but the housing crunch would have been far smaller. In my view the price drop under NGDP targeting would have been less than half as big, and defaults would have been much less than half as big under 5% NGDP targeting.

    Jon, You said:

    “I think this is very uncontroversial. Modern policy is ‘memoryless’.

    Now what about the NGDP policy. If NGDP drifts off-trend, do you expect the Fed to be memoryless? I don’t think its necessary in the same way…”

    Memoryless is exactly the problem with modern policy. It’s why AD recently fell so sharply. If the markets expected the Fed to get back on track, inflation expectations would have risen, and this would have boosted current AD. Policy with a “memory” is called “level targeting”, and is what we need whether we target inflation or NGDP.

    Jim, Thanks, I’ll have to pick up here tomorrow. I’ll look at the article.

    Because of my home computer problems, and my desire to do a post as a favor to Bryan Caplan, I have run out of time tonight. BTW, I now have a new 32-inch monitor to replace my broken CRT monitor. Don’t try powering a big hi-def monitor with an ancient (2001) computer. Now I need a new computer. More delays coming. But I’ll try to catch up on the rest of the comments tomorrow.

  44. Gravatar of Current Current
    17. July 2009 at 01:55

    Scott: “I’m not even sure bubbles exist. It’s other people that suggest high stock prices are bubbles. You say it’s not a bubble if the high prices are justified. According to whom? You, or the consensus of thousands of very sophisticated Wall Street investors?”

    According to the investors of-course. I don’t have a more complete picture than the market, but the market doesn’t have a complete picture either.

    Scott: “Other people say 1929 was a bubble, but investors didn’t think so, and later researchers found that those prices would have been very justified if we hadn’t gone into a depression.”

    That is a circular way of looking at things. Investors try to have accurate expectations of the future, then to trade on them, that is their job. In my view the onset of the depression though demonstrates that the market did not have accurate expectations of the future. If they had been right they would have marked the depression into valuations.

    Does this mean the stock prices before the depression were a bubble? In my view yes. That doesn’t mean though that I think I can outsmart the stock market in prospect, only that I think it’s reasonable in retrospect.

    Current: “Certainly stock market falls redistribute wealth. More importantly though they are associated with changing expectations of future income.”

    Scott: “Yes, but stock market falls don’t cause changing expectations of future income, they reflect changing expectations.”

    Yes.

    My point is that the change in expectations must come with a reduction to present incomes. Since agents must change their purchasing behaviour from what they had planned. At a micro-economic level some markets will see greater demand than their agents had planned, some lesser demand. Since these plans will not fit together as well as they did before there will be a loss.

    The only change isn’t redistribution.

  45. Gravatar of JKH JKH
    17. July 2009 at 03:33

    Scott S. said:

    “He claimed that only the Fed can affect the total amount of bank reserves. Actually I got a roll of quarters today from the bank, and the instant I did so aggregate bank reserves fell by $10.”

    This is a point we’ve discussed previously.

    Suppose we had a bathtub between us, and you had control over the drain, and I had control over the faucet. In any discrete time period, you can drain as much water as you like. Except that it’s a finite amount of water – a reasonably realistic assumption. Your control is limited in this sense. Although your limit is finite, you could even drain more water than the tub contains at the beginning of the period. This is because I control the faucet, and I have the option of adding any amount of water that you drain even as you drain it. In any event, I control the amount of water in the tub at the end of the period, because I have the option of responding to whatever amount you drain in whatever way I choose. Even if I choose not to respond, that is an active water level decision on my part.

    Therefore, I would say that I control the level of water in the tub as measured at the end of discrete time periods.

    Apart from this, at a more technical level, but less to the conceptual point, the banking system generally maintains vault cash in excess of the amount it has recorded as a reserve credit for a particular accounting period. This is clear from Fed reports on reserves. Moreover, there is a discrete time lag between vault cash changes and the recording of vault cash reserve credits. Technically, there is no “instantaneous” effect on total reserves when you carry away your quarters. But that’s just an operational detail.

  46. Gravatar of JKH JKH
    17. July 2009 at 03:53

    Scott,

    Just saw this, which might please you:

    http://blogs.ft.com/maverecon/2009/07/what-to-do-with-the-fed/

    Wherein, he says:

    “One aspect of interest rate policy where the Fed, along with the Bank of England and the ECB, has dropped the ball is the spread between the official policy rate and the rate banks earn on reserves held with the central bank. The Fed, which started paying interest on reserves (both required reserves and excess reserves) only on October 9, 2008, initially set the rate on reserves as as the lowest targeted federal funds rate for each reserve maintenance period less 75 basis points. As the Federal Funds target rate has moved down to zero (it currently hovers between 0 and 0.25%), the spread was reduced from 75 basis points to something between zero and 25 basis points, with the interest rate on reserves at zero.

    It would obviously have been far superior to set the Federal Funds Target rate at zero and the interest rate on reserves at negative 75 basis points. That way banks would be discouraged from taking advantage of the Fed’s wide range of liquidity-enhancing facilities only to redeposit the money with the Fed as excess reserves. In mitigation it must be said that the Bank of England and the ECB are doing even worse as regards the levels of their official policy rates and the spreads over the interest rates on reserves (or deposit rate). Bank rate in the UK still stands at 0.50 percent with the deposit rate 25 basis points lower. Again, a zero Bank rate and a deposit rate of -0.75 percent or lower would make a lot more sense. The ECB does have a 75 basis point spread of its official policy rate over the deposit rate, but its official policy rate still stands at 1.00 percent, defying both logic and gravity. With the recession in the Euro Area as deep as or deeper than in the US and the UK and with price inflation already in negative territory, a zero official policy rate and a deposit rate no higher than -0.75 percent is the only rate configuration that makes sense.”

  47. Gravatar of ssumner ssumner
    17. July 2009 at 04:43

    Jim, The article is interesting. I don’t know what to say about the technique they use to estimate liquidity. It seems very arbitrary to assume the 10 to 20 year inflation forecasts are not affected by near term inflation changes. Historically I am pretty sure the inflation rate has been something close to a random walk, meaning that changes in inflation tend to persist. So my hunch is that the true liquidity premium is smaller than they estimate. But they could be right. And I may be misunderstanding their procedure.

    Even if they are right, I’d use the unadjusted number. That might seem odd, but remember that I am interested in NGDP, not inflation. My hunch is that when liquidity concerns flare up, expected RGDP growth probably falls, indeed quite likely by at least as much as the rise in the liquidity premium. So I still think the low unadjusted figures are the one’s we should be looking at. I wish we had some market estimate of NGDP, but most of the private forecasters are pessimistic, so I am pretty sure that I am on solid ground in saying NGDP growth is likely to be too low over the next year.

    Thruth, That is possible, but remember that stocks are hurt by either extreme. Real stock prices tend to fall both during deflation, and very high inflation (because of the inflation tax on capital.) I definitely agree that things were getting gradually worse, but I still think the big mistakes were in late 2008.

    123, You’re still not seeing my point. But perhaps I am wrong, so let me try again and see what you think. I claim that the models used by people like Shiller include the anti-bubbles. So if he says the average P/E is say 13, what might be going on is that the P/Es in anti-bubbles are 9 and the P/Es in normal times are 16. In that case he might assume 16 is bubble-like, whereas it is actually rational, given good macro policy. Indeed in a sense I am arguing this has already happened to some extent. Monetary policy did get better in 1982-2007 (The Great Moderation.) And P/Es rose to reflect this. Shiller thought they were too high in 1996, but he may have been wrong. And I am saying we can even to much better than 1982-2007.

    My interest in China is more based on supply-side improvements. Their system was unbelievably inefficient under Mao. China was poorer than India and Africa when Mao died. The country has unbelievable potential. They need more economic reforms, but even what they have done so far has helped a lot. I think their monetary policymakers have done a pretty good job in this crisis by freezing the yuan. This is a de facto devaluation because Chinese productivity rises much faster than in America. But the real story there is the supply side.

    Statsguy, Yeah, It’s amazing that people are worried about inflation. As Hawtrey said in the 1930s, it’s “crying fire, fire, in the midst of Noah’s flood.”

    Rob, If you retrospectively single out two periods as being the best in American macro history, then by definition they will be followed by worse times. Take the longest win streak in football history. If you bet on the Patriots the first game after that win streak ended, you would have lost for sure. On the other hand if you bought stocks in 1926 or 1996, and held long term, you would have done OK. And if you bet on the Patriots in the middle of their win streak you probably would have done OK.

    Also, 2000’s market was like a soapy liquid surface that is gradually expanding in a spherical topology.

    Current, Your definition of a bubble is completely different from the standard definition used by people like Shiller. In their view the bubbles were when prices were irrational ex ante, when the prices were too high given the optimal forecast of the economy. I am using this standard definition.

    I don’t understand this at all:

    “My point is that the change in expectations must come with a reduction to present incomes. Since agents must change their purchasing behavior from what they had planned. At a micro-economic level some markets will see greater demand than their agents had planned, some lesser demand. Since these plans will not fit together as well as they did before there will be a loss.”

    This would imply that current income would fall even if expected future income rose.

    JKH, Your interpretation may be right, but doesn’t rescue his post at all. Let’s suppose you are right, and after imposing a penalty rate on ERs the Fed injected enough money into the banking system to keep total reserves constant (and by the way they prabably would not do this.) In that case, the Fed would have to inject enough money to make RRs go up more than 10 fold, as ERs would, by assumption, fall to nearly zero, and ERs are more than 10 times RRs right now. But this could only occur if bank deposits rose more than 10 fold! In other words, to keep the total reserve level constant the Fed would have to create hyperinflation. But that supports my point that a penalty rate on ERs would be highly expansionary. Indeed, I have said many times that a penalty rate on ERs would have to be accompanied by a big drop in the monetary base, in all probability, because it would be so expansionary.

    Thank you very much for the FT article.

  48. Gravatar of JKH JKH
    17. July 2009 at 05:11

    Scott,

    You make a very interesting point that I’ve missed … sorry, it’s probably because I haven’t been reading your blog (or any others for that matter) very closely recently.

    But it does raise another question. You make the point that the desired behaviour one might expect from the banks as a result of negative interest rates on reserves would lead one to expect the Fed to reduce the level of excess reserves as a result. That makes perfect sense to me as you describe it.

    The point that I’ve made in the past is that the Fed is actually using the excess reserve position mostly as a source of funding for its “credit easing” activities. These actions have forced the Fed to expand its balance sheet, not as a result of intentionally targeting a larger excess reserve position, but as a result of supplying specific credit oriented supported that necessarily resulted in balance sheet expansion, including a large excess reserve position as a source of funding. I’ve always thought that the Fed was paying interest on reserves essentially because of this order of priorities in the reasons for their balance sheet expansion. Paying interest on reserves is consistent with their view that the creation of the excess reserve position was actually a secondary consideration in their asset-liability strategy. I know that conflicts with the way in which you think about what they should be doing, but that is how I think they have viewed it, so far.

    So, as an observation, your expectation that the Fed would logically shrink excess reserves as a result of implementing your recommendation on reserve interest conflicts with how they have prioritized asset expansion rather than excess reserve expansion as the driving force for balance sheet expansion. They don’t want to be in a position where they are logically forced to shrink excess reserves, because they are relying on excess reserves as a source of funding for specifically targeted asset expansion activities.

    Does that make any sense, at least as an observation?

  49. Gravatar of Current Current
    17. July 2009 at 05:34

    Scott: “Your definition of a bubble is completely different from the standard definition used by people like Shiller. In their view the bubbles were when prices were irrational ex ante, when the prices were too high given the optimal forecast of the economy. I am using this standard definition.”

    Well I certainly don’t think a boom that was obvious ex ante occurred, or is likely to ever occur. But, if this is your definition of bubble then why do you talk about creating a perpetual bubble? Surely you must be using a different definition there.

    Current: “My point is that the change in expectations must come with a reduction to present incomes. Since agents must change their purchasing behavior from what they had planned. At a micro-economic level some markets will see greater demand than their agents had planned, some lesser demand. Since these plans will not fit together as well as they did before there will be a loss.”

    Scott: “This would imply that current income would fall even if expected future income rose.”

    Yes.

    Let’s suppose that Blueland is planning to invade Brownland. The army of Brownland is poorly trained and will probably lose the war.

    As a result the interest rate is very high. Nobody thinks too much about the future because believe it to be largely out of their hands.

    What happens in this case? Well, consumers are likely to spend their money. “Eat, drink and be merry for tommorrow we die.” Businesses are unlikely to invest in new plants, there would be little point. Note that both groups are acting this way because of their expectations of the future.

    Now, suppose that Brownland are surprisingly victorious in the war. If that happened then the expectations of the future would rise greatly. People would begin saving their money and businesses would begin investing. However, the structure of production would be directed towards consumptions. Those in the business of eating, drinking and merriness would have overcapacity. Those in the business of building new industrial plant or designing new products would be out of business, those businesses would have to be resurrected.

    It is quite possible in this situation that GDP would temporarily go down. Of course it would rise later.

    However, what I’m describing is quite unlikely. Expectations of future incomes though don’t normally rise very quickly. Perhaps they have in situations like that I describe, but those have been very rare.

    The opposite situation though is quite likely. So the same sort of problems of change are likely to occur in that situation.

  50. Gravatar of 123 123
    17. July 2009 at 12:48

    By eliminating monetary shocks we would reduce equity risk premium, but we will not eliminate it because fiscal, political and foreign policy risks would still remain. Even if we could eliminate abovementioned risks, average p/e ratios at 1996 levels are only consistent with implausibly low average treasury bond yields, implausibly low average corporate bond yields and corporate default rates. You are imagining a risk-free business world.

    When discussing supply side improvements, Warren Buffet likes to mention 1901-1920 USA – a period with huge supply side growth and terrible stockmarket returns.

    I agree that stopping the appreciation of yuan was a good step. What is your opinion about other China macro facts – deflation, explosive loan growth, falling exports?

    china – deflation – credit channel

  51. Gravatar of rob rob
    17. July 2009 at 13:32

    You don’t need to respond to this but I do want to amplify my point, which is why must the VERY worst of times follow the VERY best? admittedly the end of a winning streak must result in a loss, but it musnt result in a horrible losing streak? what im driving at is that you are a bit wishy-washy on whether a certain characteristc of soapy liquids matters..
    .

  52. Gravatar of ssumner ssumner
    18. July 2009 at 07:19

    JKH, Your observation makes perfectly good sense, and i have two responses.

    1. I simply don’t know enough about banking to have an intelligent opinion about whetehr these liquidity injections have helped more than some other technique, such as injecting capital. So le’t just set this issue aside and assume the fed was right to want inject another 800 billion into the banking system, and also that thye did not do so for reasons of increasing ERs but rather to help the banking system.

    2. I would still favor a different approach. If you are right then the Fed should set some arbitrary line that is consistent with the amount of reserves they injected for liquidity reasons, and then charge a penalty rate above that level, Thus suppose the Fed thinks that the banking system needed a liquidity injection large enough to move total reserves up from 80 billion to 800 billion. Then charge a penalty rate on any reserve holdings more than 10X the reserve requirement.

    You have to think at the margin here. As long as monetary policy has traction at the margin, it doesn’t matter how much reserves are hoarded. The Fed can always do more. If needed reserves are 800 billion, then the Fed can have a penalty rate on holding above that level and then put 900 billion out there. The last 100 billion will start the money multiplier process.

    Current, I am talking about creating a bubble from the perspective of Shiller’s definition. Again, I don’t think the whole concept makes much sense, but since others do, I am using their concept. Most people seem to define stock bubbles as P/E ratios above a certain threshold. (Yes, I know it is more complicated, but that is the thrust of their argument.)

    It is theoretically possible that higher future expected growth would reduce current GDP. But for various reasons I think the opposite is much more likely in the real world. To begin with, higher future expected growth tends to raise current AD.

    123, Suppose you are given a choice on January 1 1996. Invest in one of the following three and hold for either 5, 10, 20 years

    1. Stocks,
    2. T-notes
    3. Corp bonds.

    Haven’t stocks outperformed on the 5 and 10 year windows, and aren’t they likely to do so on the 20 as well? I have to admit that I don’t following his closely. But I felt that my 401k has done pretty well over the last couple decades despite a couple crashes. It may be because I hold some Asian funds, I haven’t looked at the US numbers close enough to tell.

    My other question relates to the studies that show stocks have dramatically outperformed T-bonds long term. I guess those studies are now being challenged, but I had sort of assumed that the implication of those studies was that P/Es had been too low for most of the 20th century. Are you saying the dividend yield plus the expected capital gains on stocks in 1996 was less than the yield on T-bonds? And is your comparison ex ante or ex post?

    rob, I see your point, although it always depends on which points in time you choose for comparison. Also, The post-1929 economic performance was horrible, but the post-2000 performance was more a garden variety recession, followed by a strong recovery in stocks. Remember, the US averages a recession every 4 or 5 years, so it’s not evidence of some disaster. I don’t think those soapy spheres had any role in causing the subsequent movements in GDP, as is explained in some earlier posts on did on the Great Depression.

  53. Gravatar of 123 123
    18. July 2009 at 08:31

    Stocks have massively underperformed over 7 and 13 year periods from 1996. And most likely in 2016 the performance gap will not yet be closed. I’m afraid you have to use 50yr+ horizons to justify 100% allocation to stocks in 1996.

    Ex ante in 1996 stocks are very poor investments relative to T-notes and corporate investment grade bonds. To say otherwise you have to assume equity risk premium very close to zero, or you have to assume very high expected dividend growth rate.

  54. Gravatar of ssumner ssumner
    18. July 2009 at 08:32

    123, I forgot your China question, I haven’t followed Chinese deflation. What is the rate now? But if there is deflation, it shows they were right to stop the yuan’;s appreciation, and recall that I once said that it isn’t impossible that the world might actually benefit from a weaker yuan. Once again this week really strong growth news out of China was associated with a Wall Street rally. Now obviously there were many factors (like earnings) but go back to March when people were very pessimistic about the whole world economy. Surely the backdrop of strong growth in Asia helps things at least a bit. Not just through the obvious export channel, but more importantly through the much less obvious Wicksellian equilibrium real rate channel. Asia is pushing that real rate slightly higher in world capital markets, making monetary policy at the zero bound slightly more expansionary (but still not as much as I would like.)

    China’s growth story is less export led than many over here envision. Domestic growth has always been strong (infrastructure, real estate, services, etc.) I am not happy to see the big state firms taking the lead, but hopefully those firms are (and will continue to be) becoming gradually more commercially-oriented over time.

  55. Gravatar of JKH JKH
    18. July 2009 at 12:06

    Scott,

    That makes a lot of sense to me. I’m suggesting that there is a rationale for the Fed’s excess reserve expansion that may be quite distinct from the rationale for more aggressive promotion and use of reserves in the “money multiplier” mode. What you are suggesting in effect is that the Fed could specify a “modified” required reserve level that could take this first rationale into account, and then “goose” the system with additional negative interest reserves at the margin. That to me is a perfect way of marrying what they have been thinking to date (I think) with what you would like them to do. It is operationally consistent and seamless at the meeting point between the two purposes.

  56. Gravatar of 123 123
    19. July 2009 at 09:24

    Stocks have massively underperformed over 7 and 13 year periods from 1996. And most likely in 2016 the performance gap will not yet be closed. I’m afraid you have to use 50yr+ horizons to justify 100% allocation to stocks in 1996.

    Ex ante in 1996 stocks are very poor investments relative to T-notes and corporate investment grade bonds. To say otherwise you have to assume equity risk premium very close to zero, or you have to assume very high expected dividend growth rate.

    China’s CPI fell 1.7%yoy in June, PPI fell 7.8%yoy, retail sales in June rose 15% from a year earlier – an intriguing combination of statistics.

  57. Gravatar of ssumner ssumner
    19. July 2009 at 18:00

    JKH, I agree, now if we only can get the Fed to listen. BTW, I am getting some slightly inside information that is making me more convinced than ever that the Fed didn’t think this through clearly. If I am able to release it at some point, I will. Nothing earth-shaking, but nonetheless revealing.

    123, I just don’t get what you are doing there. Why are you allowed to (negative) cherry pick the worst years? If stocks outperformed T-bills on all horizons then literally no one would ever buy T-bills. I took normal 5, 10 and 20 year horizons and you picked 7 and 13. Who’s to say which is more relevant? If we were having this debate in 2007 when the Dow was 14,200, would you still have said 6,400 was too low in 1996? Obviously not. One can never pin this down because it’s always a moving target. If you told me I had to hold for 25 years (not 50) I’d still might take stocks over T-notes in 1996. Corporate bonds is more debatable.
    But I may be off because I focus on Asian stocks, which have outperformed the US market recently. So I’m doing OK even now. Since I’ve added stocks continuously after the Asian crashes in 1997 and 2001, I bought a lot at very low levels.

    Thanks for the inflation numbers for China. I hope it goes without saying that all my macro analysis for the US must be adjusted for China. They can have a bit of deflation, and yet still have very strong NGDP growth. That’s why I have always felt a Japanese-style liquidity trap is unlikely for China. They have an amazing supply-side (albeit only because their economy was so inefficient under Mao.

  58. Gravatar of 123 123
    20. July 2009 at 07:56

    5 and 10 year horizons are no more (and no less) relevant than 7 and 13 year horizons. But let’s focus on ex ante calculations. They are very simple, we need only these things:
    1. Dividend yield in 1996.
    2. Expected growth rate of dividends in 1996.
    3. Expected future real interest rates and estimate of equity risk premium (these figures give you expected dividend yield on exit).
    And the result is that stockmarket was overvalued in 1996.

    Have you ever seen another country with negative 7% PPI and 15% retail sales growth?

  59. Gravatar of 123 123
    20. July 2009 at 09:36

    I would like to propose a more complete name for this blog post:

    “The Fed should create the mother of all stock bubbles by permanently lowering real interest rates”

  60. Gravatar of Current Current
    20. July 2009 at 10:14

    Scott: “I am talking about creating a bubble from the perspective of Shiller’s definition. Again, I don’t think the whole concept makes much sense, but since others do, I am using their concept. Most people seem to define stock bubbles as P/E ratios above a certain threshold. (Yes, I know it is more complicated, but that is the thrust of their argument.)”

    Ok, I see. I certainly don’t agree with that sort of thing.

    Scott: “It is theoretically possible that higher future expected growth would reduce current GDP. But for various reasons I think the opposite is much more likely in the real world. To begin with, higher future expected growth tends to raise current AD.”

    It is a “paper tiger”. Since future growth expectations hardly ever jump.

    The important part though is the other side. If there is a crisis which involves real losses then the economy must firstly bear those real losses. Secondly, it must also bear the cost of reallocation of resources to new production plans which involves resource and human unemployment. Thirdly, there will be a fall in aggregate demand due to the problems involved in the above. It is only this third effect that is really avoidable.

  61. Gravatar of Current Current
    20. July 2009 at 10:17

    123: “I would like to propose a more complete name for this blog post:

    ‘The Fed should create the mother of all stock bubbles by permanently lowering real interest rates'”

    What Scott is really saying is “I think the Fed should permanently change the institutional environment in a way that I think will be beneficial”. This is really what every business cycle theorist is saying.

  62. Gravatar of ssumner ssumner
    20. July 2009 at 17:23

    123, Obviously the investment community in 1996 didn’t think much of the formula you provided. Or else they had different expectations for dividend growth. I’m not saying it might not have been the right formula ex post, but markets are driven by ex ante forecasts.

    I’m not as surprised as you are by the Chinese numbers. Start with the fact that I think inflation is pretty meaningless concept in macro, it’s just some arbitrary number statisticians dream up. Second, I think nominal shocks are properly measured with nominal GDP. Third, I’m guessing that NGDP is up 6% to 8% recently in China. Fourth, it doesn’t shock me that a country with 6% or 8% NGDP growth could experience 15% retail sales growth, especially if the government is offering tax rebates to buy cars that will expire soon. (Which I think is true, but don’t quote me because I am not certain.)

    Current, Yes only the third part is avoidable, but in this recent crisis 90% of the damage came from the third part (post-August 2008.)

  63. Gravatar of 123 123
    21. July 2009 at 07:45

    If we disregard sunspot stockmarket cycle theory, then every longterm stockmarket forecast can be expressed in terms of the formula I provided. In the late 90s there was both a talk of permanent productivity miracle (which somehow should increase dividend growth rate), and of zero equity risk premium (remember those Dow 36,000 guys). In 1996-2000 market implied dividend growth rate got so crazy and market implied equity risk premium got so out of the line from the actual risk appetite of investors that a big bear market followed.

    Improved Fed policy should lower the equity risk premium. Will it have any efect on long term real risk-free interest rates?

    Nominal gdp of china has increased by 4% yoy in second quarter, indicating at least some nominal disturbances.

  64. Gravatar of ssumner ssumner
    22. July 2009 at 05:57

    123, I don’t know enough finance to know whether less systemic risk lowers or raises the risk free rate. But I’d guess that it would raise it.

    I’m surprised by the NGDP number. Have they really had 4% deflation in their GDP deflator YOY? I knew China was a pretty flexible price economy, but that’s amazing. I also read that RGDP rose at a 16% annual rate in Q2. (China reports YOY numbers as you obviously know.) Talk about “green shoots!”

    I agree it was a pretty big (negative) NGDP growth shock. But notice it was still positive, which helped China avoid the liquidity trap/falling nominal wage problems of countries that saw their NGDP fall sharply. I’m certainly not claiming China follows NGDP targeting. And I don’t know how much the usual Phillips curve model is affected by the heavy government role in bank lending.

  65. Gravatar of 123 123
    23. July 2009 at 10:10

    So if we reduce systemic risk we have higher real risk free rate and lower risk premium – this combination does not really imply a permanent stock bubble.

  66. Gravatar of Scott Sumner’s Stock Market Theory | Austrian Economics Blog Scott Sumner’s Stock Market Theory | Austrian Economics Blog
    23. July 2009 at 12:47

    […] Scott Sumner’s Stock Market Theory BY admin ON July 23rd, 2009 Scott Sumner, who’s “Nominal GDP (NGDP) target”-theory I analyzed here have a post with a misleading and (presumably deliberately) provocative title “The Fed should create the mother of all stock market bubbles, permanently”. […]

  67. Gravatar of ssumner ssumner
    24. July 2009 at 05:26

    123, You have to go back to my original definition, which was purely in terms of P/E ratios. My claim was that it would produce higher average P/E ratios, and thus that it would look like a bubble to those who use P/E ratios to determine bubbles. Look, I don’t even believe bubbles exist, so obviously I agree that it would not really be a bubble, taking into account all the issues you raise.

    But perhaps you are getting at something different. Would a Shiller-type model immediately detect what had occurred? Or would they only gradually ascertain over time that the risk free rate had risen? My view is that it would be the latter, as Shiller was very skeptical in 1996, when we had a little taste of what I am hypothesizing here. Investors in 1996 (and 2006) were starting to buy into the Great Moderation hype. Now it turns out that was premature in both years. But what if we do actually get there? It’s not that far-fetched as one could argue that Australia has already achieved this goal. I think that it would initially look like a bubble to people like Shiller, Krugman, Roubini, et al.

  68. Gravatar of 123 123
    24. July 2009 at 11:52

    I believe that increased risk free rate would decrease the average P/E rate.

  69. Gravatar of ssumner ssumner
    24. July 2009 at 17:14

    123, I misread your previous comment. I thought you were saying stocks would be higher, but it would be no bubble. So you are saying stocks would not be higher. Why wouldn’t less systemis risk move people from safe assets like T-bonds into riskier assets like stocks?

  70. Gravatar of 123 123
    24. July 2009 at 23:33

    While I believe that lower systemic risk would raise the average stock prices a bit (though not to 1996 levels), this belief is not directly supported by theory. Lower systemic risk would have two offsetting effects – lower risk premium and higher risk free rate. The price of treasuries would fall to attract people. Stocks would rise because of lower risk premium and would fall because treasuries will be more attractive.

  71. Gravatar of ssumner ssumner
    25. July 2009 at 06:22

    123, Isn’t that flawed reasoning to say:

    1. Treasuries attract less demand
    2. Therefore the price of Treasuries falls
    3. Therefore people switch from stocks into the newly cheaper Treasuries.

    In standard “related goods” theory of demand in microeconomics, when demand for good A falls, demand for it’s substitute rises, and the price of its substitute also rises.

    Maybe it is ambiguous, but it seems really counterintuitive that stock prices wouldn’t rise. It seems to me that the fall in Treasuries occurs precisely because more people want stocks.

  72. Gravatar of Current Current
    25. July 2009 at 09:44

    123, Scott,

    This discussion is rather like the Rabbit in “Monty Python and the Holy Grail”. It looks innocent, but it has the bodies of five score of economists strewn about it’s lair.

    Sometimes stock prices follow interest rates. Sometimes the two oppose each other. If you think about the number of variables involved in this it’s really impossible to say what will happen in a very general way.

  73. Gravatar of ssumner ssumner
    26. July 2009 at 05:14

    Current. Good point.

    BTW, The question an economist should never ask is:

    What happens if the price of X changes?
    What happens if he exchange rate changes?
    What happens if the interest rate changes?

    The the effect of higher oil prices from less supply are very different than the effect from more demand (in one case quantity goes up, in the other quantity goes down.) Instead you should ask what happens as a result of the thing that caused the price change. Maybe this is obvious, but I think it is in the background of the point you raise. My hunch is that if the initial causal factor is more steady NGDP growth, then in some way the result has to be higher equity prices. But I am not good at “general equilibrium” analysis, and thus am having trouble coming up with a convincing argument.

  74. Gravatar of Current Current
    26. July 2009 at 06:37

    Asking the question “What happens if the price of X changes?” is reasonable for some purposes. It’s an important part of explaining the market process, Hayek uses it in “The Uses of Knowledge” to discuss transmission of price changes through various markets.

    But, for many purposes it isn’t a correct to stop at discussion of price changes. This is the problem here.

    Analytically speaking the returns from shares have two components, though things can be split up in several ways. Firstly, there is the interest portion which is due to time-preference. Secondly, there is the profit portion which is due to taking risks. If you prefer you can put the effect of productivity into the interest portion or into the profit portion.

    However, this sort of analytical split doesn’t help us that much for this problem. For example, suppose that improved monetary policy means that all businesses are now taking less risk. Superficially we may say that all others things being equal the value of businesses must rise.

    We can’t say though that they will all rise equally. Some businesses will benefit more from stability than others. In particular small businesses may benefit more than large businesses. This may mean that the change in the price of stocks doesn’t represent the true picture. Large stocks could even fall because it is considered that their smaller rivals now have better business prospects. During the boom of the last 10 years in the UK the smaller cap stocks did much better than the FTSE100.

    More importantly though debt makes things much more complicated. Debt is the most tax efficient way of dealing with a steady revenue stream. So, businesses may decide that if they have a more steady income they will use it more. They could pay for the buy-back of shares by issuing bonds.

    It’s very hard to say what will happen overall.

  75. Gravatar of 123 123
    26. July 2009 at 09:43

    Scott,
    micro intuition has led me to say that most likely stockmarket will rise. However we are dealing with a full macro picture here. The key issue is how the absence of macro crises will impact the willingness to save.

  76. Gravatar of ssumner ssumner
    27. July 2009 at 04:24

    Current, You said:

    “Asking the question “What happens if the price of X changes?” is reasonable for some purposes. It’s an important part of explaining the market process, Hayek uses it in “The Uses of Knowledge” to discuss transmission of price changes through various markets.”

    I can’t let that pass. Are you saying that it is OK to start an analysis with a price change, without any reference to why prices changed? I’d like to see an example, as I can’t think of any. In fact, I may do a post on this point.

    I don’t have any problem with you other points about how complicated things are.

    123, I hadn’t thought of the saving angle. Since I have your attention, what do you think will happen to real interest rates when baby boomers like me retire? My intuition is that real rates will go down, because I want them to go up, and it seems like my generation has gotten screwed at every step of life by being so numerous (harder to find jobs, etc.) On the other hand, it would seem like boomer retirements should lower the national saving rate and boost real interest rates. Do you know what the standard view is? I have never been interested in capital theory for some reason, and I have a hunch that I am thinking about this in the wrong way. Perhaps I need a stock approach–the stock of capital should be very large just as boomers are about to retire, is that the explanation?

  77. Gravatar of Current Current
    27. July 2009 at 06:19

    Scott: “I can’t let that pass. Are you saying that it is OK to start an analysis with a price change, without any reference to why prices changed? I’d like to see an example, as I can’t think of any.”

    It depends on what is being analysed and why. This is what I was thinking of when I wrote that.

    Hayek:
    “Fundamentally, in a system in which the knowledge of the relevant facts is dispersed among many people, prices can act to coördinate the separate actions of different people in the same way as subjective values help the individual to coördinate the parts of his plan. It is worth contemplating for a moment a very simple and commonplace instance of the action of the price system to see what precisely it accomplishes. Assume that somewhere in the world a new opportunity for the use of some raw material, say, tin, has arisen, or that one of the sources of supply of tin has been eliminated. It does not matter for our purpose””and it is very significant that it does not matter””which of these two causes has made tin more scarce. All that the users of tin need to know is that some of the tin they used to consume is now more profitably employed elsewhere and that, in consequence, they must economize tin. There is no need for the great majority of them even to know where the more urgent need has arisen, or in favor of what other needs they ought to husband the supply. If only some of them know directly of the new demand, and switch resources over to it, and if the people who are aware of the new gap thus created in turn fill it from still other sources, the effect will rapidly spread throughout the whole economic system and influence not only all the uses of tin but also those of its substitutes and the substitutes of these substitutes, the supply of all the things made of tin, and their substitutes, and so on; and all his without the great majority of those instrumental in bringing about these substitutions knowing anything at all about the original cause of these changes. The whole acts as one market, not because any of its members survey the whole field, but because their limited individual fields of vision sufficiently overlap so that through many intermediaries the relevant information is communicated to all. The mere fact that there is one price for any commodity””or rather that local prices are connected in a manner determined by the cost of transport, etc.””brings about the solution which (it is just conceptually possible) might have been arrived at by one single mind possessing all the information which is in fact dispersed among all the people involved in the process. “

  78. Gravatar of 123 123
    27. July 2009 at 13:16

    Scott,

    I’ve got bad news and good news. The bad news is that demographics is important in determining expected future real interest rates. The good news is that the real baby boomers are much younger than you, you should have seen them on your frequent trips to China, their government is responsible for the savings glut.

    Demographic transition also influences stockmarket through another channel – profit margins.

  79. Gravatar of ssumner ssumner
    28. July 2009 at 06:49

    Current, Hayek is making a different point, he is saying that prices play an important role in decision-making. I was arguing that one cannot understand what is going on in a market by simply at prices. High oil prices don’t tell me whether demand has increased or supply has decreased, that is my point. But high oil prices do send me a signal to conserve, that was Hayek’s point.

    123, Yes, but doesn’t your “good news” mean that real rates will be low when I retire in 8 years? After all, the Chinese will be saving a lot of money. I consider that bad news.

  80. Gravatar of Current Current
    28. July 2009 at 07:02

    Scott, regarding prices, I see what you mean and I agree.

    Scott, 123,

    This is a very difficult topic. You have to remember that expectations play a part here too, as does the length of the process of production. I’m not sure that it is particularly certain what will happen.

  81. Gravatar of 123 123
    28. July 2009 at 09:39

    Scott,

    after 8 years real rates will be low and Chinese will be eager to buy your assets at a high price.

  82. Gravatar of ssumner ssumner
    29. July 2009 at 06:45

    Current, Your comment led me to a post, which even got Mankiw’s attention. So thanks.

    123, I suppose that’s good, but for my heirs, as I plan to live on the interest. But I see your point, if the price doubles and the rate falls in half, the the dollar flow of income is the same.

  83. Gravatar of Incoming links and arriving packages | Quotar Blog Incoming links and arriving packages | Quotar Blog
    6. August 2009 at 23:12

    […] bubble detection proposal was mentioned by Scott Sumner on TheMoneyIllusion this week, and there’s also an interesting discussion of it on Baseline Scenario […]

  84. Gravatar of Carl Lumma Carl Lumma
    9. August 2009 at 15:29

    ssumner wrote:
    >David, The mistake you make is to assume that financial
    >crises make the economy unstable. They do not. It is NGDP
    >fluctuations that make it unstable. So let’s suppose
    >everyone gets leveraged, and we have a financial crash.
    >Big deal. Under NGDP targeting it merely redistributes a
    >little wealth, the economy keeps chugging ahead.

    I think this only works if the financial sector is a small part of the economy. It’s something on the order of 25% GDP, even without the NGDP growth guarantee. You’d need a lot of helicopters. Speaking of which, in a total banking failure, how do you get the money out there to maintain 5% NGDP growth?

  85. Gravatar of ssumner ssumner
    10. August 2009 at 06:10

    Carl, You asked;

    “Speaking of which, in a total banking failure, how do you get the money out there to maintain 5% NGDP growth?

    Open marker purchases of bonds, which is the normal policy.

    And the financial sector is nowhere near 25% of GDP, I think it is below 10%.

  86. Gravatar of Carl Lumma Carl Lumma
    10. August 2009 at 13:24

    Scott: You’re right, it’s closer to 10% (13% in 2002 according to Wikipedia). I’m not convinced that in a total banking failure, resulting from the popping of the mother of all bubbles, the bond markets would function so well…

  87. Gravatar of ssumner ssumner
    11. August 2009 at 06:46

    Carl, Well obviously one could not get “the mother of all crashes” in an economy with stable NGDP growth, so I really don’t think that hypothetical is worth talking about.

    But to answer your specific question, as far as I know bond markets do continue functioning even when the market crashes.

  88. Gravatar of Carl Lumma Carl Lumma
    11. August 2009 at 09:40

    Scott: The hypothetical came from other commenters (e.g. David Pearson). If the U.S. stock market went to zero (say), I doubt anyone would want U.S. government debt (bonds).

  89. Gravatar of ssumner ssumner
    13. August 2009 at 08:03

    Carl, I would, so demand would not fall to zero. During the Great Depression when stocks lost 85% of their value, the price of government bonds actually went up. The bond market was very strong.

  90. Gravatar of スコット・サムナーのマネー観、マクロ観 by Scott Sumner – 道草 スコット・サムナーのマネー観、マクロ観 by Scott Sumner – 道草
    7. May 2011 at 23:59

    […] どうしてそれで資産価格が安定化するのか?ここも私が同僚たちと違うところだ。ウッドフォード流のフォワードルッキングなモデルをマネタリスト風にしたものを想像してみよう。伝統的なマネタリズムと異なり、我々はウッドフォード流に金融政策の期待将来経路の変化を重視する。マネタリズムの流儀によって超過現金バランスの伝達メカニズムを前提にしているが、これは長期の場合に限る。つまり貨幣供給の拡張は、それが恒久的だと期待されていれば超過現金バランス効果を通じて期待NGDPを引き上げるということである。同様に、それは株式やコモディティや不動産のような資産価格を直ちに上昇させる。そして同様にその時点のADを引き上げるだろう。これは賃金が硬直的であるからで、これら資産価格の上昇が企業の資産やコモディティや不動産からの産出を増加させるからである。そして雇用と富が増加するから消費も増える。このように、ウッドフォードの議論と同じく将来のNGDPに大きな影響を与えると期待される政策は、その時点のNGDPにもまた強力な影響を及ぼす(邦訳)。では政策のラグの問題はどうだろうか。   […]

Leave a Reply