Is there an upper bound on long term interest rates?

No, the central bank can raise inflation and interest rates as high as it likes.  But people who worry about central bank ineffectiveness seem to have something else in mind.  Here’s Izabella Kaminska:

And yet despite all that additional supply, longer-dated bond yields remained suppressed “” fueling Greenspan’s famous conundrum. Many theories have been proposed as to why that happened, though in our opinion Ben Bernanke’s “savings glut” explanation is one of the most compelling “” especially when tied to China’s rampant (currency-depreciation linked) Treasury purchases at the time.

Given what we’ve all come to know about outright bond purchases by central banks, one can safely conclude that these moves injected liquidity directly into the US system, suppressing yields. In fact, one might even say, it was a type of stealth QE operation by the Chinese (and the Japanese), focused on stimulating American demand for their own export-led economies. The side-effect “” the accumulation of all those Treasuries “” however, also presented these central banks with the ability to challenge the Fed’s monopoly power over the Treasury market and to quasi dictate rates themselves.

In one respect, the Fed was desperately trying to raise rates “” while Asian asset purchases were continually undermining those attempts. The Fed had control over the front end of the yield curve, but China (and other Treasury-consuming nations) were increasingly dictating the yields at the back of the curve.

Coupled with demand from western pension funds and asset managers, there was simply not enough safe US Treasury assets to go round. The more the Fed tried to raise rates, the more the yield curve inverted.

This is an interesting example of what happens when people confuse money and credit, more specifically easy money and easy credit.  The two concepts are actually unrelated, but are often confused.

The Fed can always raise short term rates.  It’s true that this doesn’t always result in higher long term rates.  But that’s not a sign that monetary policy is ineffective, rather exactly the reverse.  The Fed raises short term rates to keep inflation down close to 2%.  If long term rates don’t budge very much, that’s a sign that monetary policy is credible.  In the 1960s and 1970s both long and short rates tended to rise together, hence increases in short rates weren’t enough to get ahead of the curve.  Real rates didn’t rise, as the Fed wasn’t following the Taylor Principle.  Because the “tight money” policies weren’t credible, inflation expectations rose.  The final sentence of the quotation is quite revealing.  An inverted yield curve doesn’t indicate failure, rather it’s exactly what the Fed wants to occur when it tightens money, it wants to reduce expectations of NGDP growth, and an inverted yield curve indicates it did just that.

The comments on China and Japan are also peculiar.  Kaminska is claiming that increased government saving is expansionary for aggregate demand.  I actually find that a refreshing change from the ordinary Keynesian argument that more saving (from the public or government) is contractionary.  Both views are wrong, monetary policy determines the path of NGDP, not saving.  Yes, saving can reduce velocity, but that would only reduce AD if the Fed had a money supply rule.  Under any sort of modern inflation targeting or Taylor Rule the Fed offsets declines in V with more M.  So saving is not contractionary.  But it’s also not expansionary.  Kaminska should have said that government saving boosts exports, not that it boosts demand in export-led economies.  For every extra dollar of exports, there is one less dollar of domestic expenditure (mostly less consumption in the case of China.)

And there was no “stealth QE” for the US economy, as we weren’t at the zero rate bound back then.  Almost everyone agrees that the Fed drives NGDP in normal times, when interest rates are above zero.  Credit is one thing, money is another.  Yet for some reason lots of very smart people get them confused.  (Although Milton Friedman never did–recall he said that ultra-low interest rates are usually a sign that money has been tight.)

Part of the problem is terminology.  We throw around terms like “spending” in very vague and misleading ways.  Some people mean consumption, others mean a reduction in money demand.  Again, the two concepts are completely unrelated.  We need more spending in the sense of less money hoarding, and we (in America) need less spending in the sense that we need more saving.

HT:  Tyler Cowen

PS.   The saving conundrum did not cause any problems with Fed policy during the housing boom—they were able to produce roughly the sort of growth in NGDP that they wanted.  If they thought it was too high, they would have raised short term rates even higher.

PPS.  If you disagree with my comment about Asian government saving and AD, check out the growth in Japanese NGDP since 1993—it’s zero.


Tags:

 
 
 

23 Responses to “Is there an upper bound on long term interest rates?”

  1. Gravatar of Nick Rowe Nick Rowe
    22. December 2011 at 14:10

    What a coincidence! I’ve just spent the last couple of hours puzzling over that exact same article, trying to figure it all out. And I eventually came to the conclusion that finance people and money/macro people just don’t speak the same language. Basically, they don’t get the monetary policy/General equilibrium perspective.

    On the other hand though, I think that they understand a lot of stuff that we don’t understand. In particular, what exactly is the role of “safe assets” (e.g. US Tbills) in the payments system, and as colateral? Why exactly is a “shortage” of “safe assets” a problem? And what exactly does a “shortage” mean? Does it mean excess demand?

    Is a shortage of safe assets like a shortage of coffee? (Doesn’t matter except for coffee drinkers).

    Or like a shortage of medium of exchange? (Disrupts all trade.)

    Or something in between, like a shortage of shopping bags? (Disrupts some trade.)

  2. Gravatar of ssumner ssumner
    22. December 2011 at 14:31

    Nick, They certainly understand plenty of things that I don’t understand. But given they don’t understand monetary economics, can we have any faith in what they say about shortages of T-bills?

    Is there anyone on Earth who understands both money and credit markets? Is the human brain capable of encompassing both realities?

  3. Gravatar of 123 123
    22. December 2011 at 14:47

    Scott, Nick,

    Shortage of safe assets is like a shortage of coins in some Latin American countries (it is a shortage of medium of exchange that disrupts some trade). In Latin America small retail transactions are disrupted, in the US large financial transactions are disrupted.

    T-bills have important role as a medium of exchange, and repo should be included in M3. There is no money-credit dichotomy.

    As a regular reader of FT Alphaville, I really hope Izzy will read this blog. Nick spent a couple of hours today, and I have to spend almost two minutes every day when Izzy writes about macro. Izzy’s conclusions look like they are generated by a random number generator.

    A couple of days ago I spent two hours wrestling with Bill Gross’s FT piece, which was puzzling in almost the same way Izzy’s writings are. But Gross is not Izzy, and those two hours were well spent: http://www.themoneyillusion.com/?p=12336#comment-122262

  4. Gravatar of 123 123
    22. December 2011 at 14:49

    No disrespect to Izzy was intended in my previous comment. As Cowen said, she raises really important questions.

  5. Gravatar of Nick Rowe Nick Rowe
    22. December 2011 at 15:26

    Scott: “Is there anyone on Earth who understands both money and credit markets? Is the human brain capable of encompassing both realities?”

    I’ve been wondering the same.

    I get the feeling that Brad DeLong might be a good candidate. He understands money. He understands more about finance than most. But I was reading his recent post too, on the same subject, and I can’t quite figure it out.

    http://delong.typepad.com/sdj/2011/12/show-me-the-collateral-blogging-one-of-the-three-major-market-failures-underlying-our-current-difficulties.html

    123: your comment cheered me. Made me feel less thick.

    “T-bills have important role as a medium of exchange, and repo should be included in M3. There is no money-credit dichotomy.”

    If so, then I understand why a shortage of Tbills matters. I still don’t fully understand the manner in which they function as a medium of exchange though, but I’m working on it. And, why can’t large bundles of banknotes do the same job? (Or, for convenience, the bonds issued by a bank with 100% reserves.)

    And, if all that’s needed is colateral, why not just use bundles of common stocks, with a (say) 50% (or more) haircut? There’s no shortage of common stocks.

  6. Gravatar of 123 123
    22. December 2011 at 15:57

    Nick:

    “If so, then I understand why a shortage of Tbills matters. I still don’t fully understand the manner in which they function as a medium of exchange though, but I’m working on it.”
    Nick, you recently blogged that a recession is not a drop in output. It is a drop in the ability of to exploit gains from trade in newly-produced and old things. A next step is needed. Recessions happen when it is harder to exploit gains from trade in goods AND financial assets. T-bills and repo lubricate trade in the financial assets.

    “And, why can’t large bundles of banknotes do the same job? ”
    Yes they can. But Bernanke did not print enough of the banknotes – that’s why we have a recession. And repo is cheaper than the printing press – that’s why it is a bad idea to print large bundles of banknotes.

    “And, if all that’s needed is colateral, why not just use bundles of common stocks, with a (say) 50% (or more) haircut? There’s no shortage of common stocks.”
    This is a good idea, but regulatory and technical barriers prevent us from fully exploiting it. By the way, I remember that some time ago you discussed cows as a collateral in your blog – that was a good idea too, but I’m afraid Frankendodd bill will not allow this. On a more serious note, during the last meeting ECB extended the list of collateral eligible for monetary policy operations – the conclusion is that ECB agrees that recessions are failures to exploit gains from trade in goods and financial assets.

  7. Gravatar of Nick Rowe Nick Rowe
    22. December 2011 at 17:34

    123: “A next step is needed. Recessions happen when it is harder to exploit gains from trade in goods AND financial assets. T-bills and repo lubricate trade in the financial assets.”

    Yep. I had already (mentally) made that next step. It’s the third sentence in your above quote I’m stumbling over. I need to understand better *how* they lubricate trade. And why Tbills can do that job better than other assets.

    But I’m getting there. Your comments are helping me. You are speaking to me on my wavelength. But I’m not quite there yet.

  8. Gravatar of Mike Sax Mike Sax
    22. December 2011 at 17:52

    “We need more spending in the sense of less money hoarding, and we (in America) need less spending in the sense that we need more saving.”

    If you could elaborate the difference a little it would be helpful. Thank you

  9. Gravatar of ssumner ssumner
    22. December 2011 at 18:56

    123, I have no problem including T-bills with M3, but M3 is credit, not money. It’s monetary policy that determines NGDP, not the amount of T-bills in circulation.

    Nick, DeLong knows a lot more about finance than I do, that’s obvious. But just between you and me I think we know a bit more about money. Still, you may well be right about him have the greatest total knowledge in both fields.

    Recently I have the feeling he’s more monetarist than I am. He keeps talking about the need to print money, and I want the Fed to reduce the public’s demand for money (by setting a higher NGDP target.) He wants that too, and I’m also fine with printing money, but he has a slightly more monetarist emphasis. On the other hand he includes T-debt, which I don’t–that’s definitely Keynesian.

    Regarding your response to 123, A few years ago Australia had virtually no national debt, and they have no AD problems. I don’t think public debt is the issue, monetary policy is.

    Mike, We need to reduce the demand for base money, or increase the supply. That has nothing to do with less saving. A bank can reduce it’s excess reserves by swapping them for T-bills. No dissaving has occurred, but there is less demand for base money, which is expansionary. (Obviously I’m talking about normal times here, when rates are above zero. I making a general observation, not one specifically aimed at the zero bound problem.)

  10. Gravatar of Robert Olson Robert Olson
    22. December 2011 at 21:22

    Scott,

    Can you really blame her for thinking monetary policy is ineffective when she’s just taking a Greenspan quote? To be fair she’s misconstruing credit with money, but to give her credit she’s misunderstanding it through asset purchases. She’s suggesting that because China controls a vastly larger stock of assets than the Fed, it actually has greater control over monetary policy than the Fed.

    That’s a lot better than the simple “interest rate low=loose money.”

    Can the Fed actually reduce NGDP growth if China insists on buying US bonds? I suppose not, since they’d have to use US dollars to buy them, which they’d earn from exports, and then they are really just reinvesting US dollars back into the US economy. They really are just easing some forms of credit, perhaps especially ultra-safe assets if the Chinese are risk-averse.

    Monetary economics is somewhat challenging for us simpletons. I may be totally off here.

  11. Gravatar of ssumner ssumner
    23. December 2011 at 06:51

    Robert, You said:

    “She’s suggesting that because China controls a vastly larger stock of assets than the Fed, it actually has greater control over monetary policy than the Fed.”

    It’s not about who controls the biggest stock of assets, it’s who controls the money supply. That’s the Fed.

    The Fed can always reduce NGDP growth, I don’t think anyone seriously disputes that.

  12. Gravatar of flow5 flow5
    23. December 2011 at 07:27

    Some definitions (money & credit): Money school advocates contend that the money stock, in a trend sense, should keep pace with the economy. The credit school adherents believe the statistical gudelines of monetary mangement should be more broadly based. They rely on such statiscal criteria as the volume and rate-of-change in commercial bank credit; total, requied, excess, nonborrowed, and free or net borrowed reserves. In our monetary system (with immaterial exceptions), it may be said that as time deposits grow the primary money supply shrinks pari passu – unless offset by an expansion of bank credit. I.e., time deposits rather than being a source of loan funds are the indirect consequence of prior bank credit creation.

    Savings: If savings are not spent on some form of consumption (dis-savings) or are not spent on some form of investment, their velocity or turnover is zero. If savings are impounded within the commercial banks they are lost to investment, exerting a depressing effect on prices, production, employment, & incomes.

    Interest rates: Keynes’ liquidity preference curve is a false doctrine. The Keynesian training of the FED’s technical staff advises them that interest is the price of money, not the price of loan-funds. They therefore decided that the money supply could be controlled through the manipulation of interest rates (the theory that there is no difference between money & liquid assets). Thus they fail to recognize the crucial importance of the difference between money creating institutions and financial intermediaries.

  13. Gravatar of 123 123
    23. December 2011 at 07:52

    Scott, You said:

    “I have no problem including T-bills with M3, but M3 is credit, not money. It’s monetary policy that determines NGDP, not the amount of T-bills in circulation.

    The questions “what is money?” and “who is responsible for monetary policy” are different. Cigarettes were money in Nazi concentration camps, and treasury was responsible for monetary policy 20 years ago in the UK.
    Milton Friedman said M3 is money.

    You said:
    “Regarding your response to 123, A few years ago Australia had virtually no national debt, and they have no AD problems. I don’t think public debt is the issue, monetary policy is.”

    Newsflash – the real reserve requirements these days are set by Basel, and T-bills count as reserves under Basel liquidity regulations. As Australia has got almost no sovereign debt, one month ago Izzy reported that Australian central bank has decided to create a CLF (committed liquidity facility), otherwise the lack of T-bills could produce a monetary contraction:
    http://ftalphaville.ft.com/blog/2011/11/25/765031/manufacturing-quality-collateral/

    Saying that Australia had virtually no national debt, and they have no AD problems, does not prove that the national debt isn’t money. Otherwise I could prove that the 500 Euro note is not money, as Australia has no 500 dollar bills.

  14. Gravatar of 123 123
    23. December 2011 at 07:58

    Nick:
    “I need to understand better *how* they lubricate trade. And why Tbills can do that job better than other assets.”

    Gary Gorton is a good source on collateral and the shadow banking system.
    Tbills have two advantages – banking regulations promote them, and as Tbills have virtually no risk of nominal value, it is very easy to use them as money. If you want to use cows as a collateral, your operating costs will be much higher.

  15. Gravatar of TravisA TravisA
    23. December 2011 at 08:06

    Thanks for the great post. I had read this article by Kaminska and thought it was very bad, but in a way that I couldn’t articulate. But you did. Whenever I read Kaminska, my IQ drops 5 points.

    Scott, a couple of questions..

    1) What is your definition of money and credit? Is money just currency + reserves? Is credit the total dollar value of all loans? Consequently, the interest rate is the price of credit?

    2) You said that a change in savings doesn’t increase or decrease NGDP if the Fed follows a Taylor rule (or more obviously targets NGDP). Does a change in savings make a difference in RGDP over the long term? Presumably, increased savings increases the capital stock which will allow for a greater production of goods in the future.

  16. Gravatar of TravisA TravisA
    23. December 2011 at 08:31

    Somewhat tangentially, you’ve mentioned before your frustration with people in the past who think that increased government spending increases RGDP but increasing the money supply by the central bank increases only inflation. You wondered what type of model those people have. I thought of something simple that might be in their minds.

    If the government borrows money to spend, it makes purchases in markets where the supply curve is relatively flat (labor markets for example). When the Fed increases the money supply, the money flows into markets where the supply curve is much more vertical (commodity markets for example). So in that way, you get a different inflation trade-off between the government and the Fed increasing AD.

    In practice, this doesn’t happen — at least it isn’t the primary impact. But I suppose it is possible.

  17. Gravatar of ssumner ssumner
    25. December 2011 at 09:07

    flow5, Don’t mix up saving and savings.

    123, I don’t follow your comments at all. Cigarettes were money in prison camps, but not in the US. So what’s the point? Try spending cigarettes or T-bills at the store.

    I don’t think T-bills count toward reserve requirements, unless the rules have recently been changed. The Fed sets reserve requirements, not Basel. Basel sets capital requirements.

    I think it’s a mistake to focus on the banking system. NGDP is determined by the supply and demand for non-interest bearing currency.

    TravisA. I define money as the base. Yes, credit is loans and the interest rate is the price of credit.

    Yes, in the long run more saving leads to more RGDP.

    You said;

    “When the Fed increases the money supply, the money flows into markets where the supply curve is much more vertical (commodity markets for example).”

    Another big mistake—money doesn’t flow into markets. And even if it did, people almost never spend base money on housing.

  18. Gravatar of 123 123
    25. December 2011 at 15:16

    Scott, my point was that the questions “what is money” and “who is responsible for monetary policy” are completely separate, and the answers can be non-obvious. That was the reason I disagreed you said: “I have no problem including T-bills with M3, but M3 is credit, not money. It’s monetary policy that determines NGDP, not the amount of T-bills in circulation.”

    “Try spending cigarettes or T-bills at the store.”
    Try spending nickels and dimes in the financial markets.

    “I don’t think T-bills count toward reserve requirements, unless the rules have recently been changed. The Fed sets reserve requirements, not Basel. Basel sets capital requirements.”
    After the Fed expanded the balance sheet in 2008, reserve requirements you have in mind are obsolete, though Plosser wants to revive them. From 2015 Basel liquidity standards that regulate the asset side of bank balance sheets will apply. 2015 is the new 1936. Please read these this:
    http://www.bloomberg.com/news/2011-06-21/basel-liquidity-rules-may-bite-harder-than-capital-thresholds.html

    “I think it’s a mistake to focus on the banking system. NGDP is determined by the supply and demand for non-interest bearing currency.”

    I don’t agree. NGDP is determined by the supply and demand for money. It doesn’t matter whether money is interest-bearing, or non-interest bearing. Of course, if money is interest-bearing, larger quantities of money are needed.
    In 2008 the shock started with the decreased supply of interest-bearing money. Low production of nickels came later.

  19. Gravatar of 123 123
    25. December 2011 at 15:20

    Another link about Basel reserve requirements (“Global regulators are targeting possible loopholes in draft rules to bolster banks’ liquidity”) :

    http://www.bloomberg.com/news/2011-12-22/basel-regulators-seek-to-close-loopholes-in-bank-liquidity-rules.html

  20. Gravatar of 123 123
    25. December 2011 at 15:23

    Another point – treasuries and other securities are reserves in the shadow banking system. The problem is that here reserve requirements aka haircuts are set by markets in the procyclical way.

  21. Gravatar of ssumner ssumner
    31. December 2011 at 08:32

    123, Saying reserve requirements don’t matter is very different from saying T-bills count toward reserve requirements.

    The slowdown in base growth occurred at the end of 2007 and early 2008, before the slowdown in other categories, if I’m not mistaken.

    I agree that money need not be created by the Fed, but in America it is. Other categories of what you call “money” matter only to the extent that they impact the supply or demand for base money. That’s up to the Fed.

    You said;

    “Another link about Basel reserve requirements (“Global regulators are targeting possible loopholes in draft rules to bolster banks’ liquidity”) :”

    Liquidity and reserve requirements are two completely separate concepts. Only base money counts toward reserve requirements.

  22. Gravatar of 123 123
    1. January 2012 at 12:57

    Scott,

    “Liquidity and reserve requirements are two completely separate concepts. Only base money counts toward reserve requirements.”
    Liquidity and reserve requirements are equivalent. Reserve requirements increase demand for base money directly. Liquidity requirements increase demand for base money and they increase demand for close substitutes of base money. Reserve requirements of x% are equivalent to liquidity requirements of y%, where the ratio of x and y depends on the supply of close substitutes for base money. As reserve requirements no longer function, it is the liquidity requirements that can create a lot of damage. So T-bills count toward reserve requirements that matter (Basel requirements). In Australia, where there is a lack of close substitutes for base money, the central bank will expand the balance sheet, otherwise Basel liquidity requirements could cause a monetary contraction. 2015 could be the new 1936.

    “The slowdown in base growth occurred at the end of 2007 and early 2008, before the slowdown in other categories, if I’m not mistaken.”

    In 2003-2007 the growth rate of monetary base decelerated along a very steady trend. This deceleration stopped in 2007.

    The decline in the growth rates of broad money was insufficiently offset by the Fed – that was the main problem.

    “I agree that money need not be created by the Fed, but in America it is. Other categories of what you call “money” matter only to the extent that they impact the supply or demand for base money. That’s up to the Fed.”

    In America the Fed controls the monetary policy, and it is up to the Fed how the changes in broad money impact the aggregate demand. But the definition “Money is the liabilities of agency responsible for the monetary policy” is unhelpful, as it obscures the important fact that the money supplied by the central bank competes with other kinds of money.

  23. Gravatar of Scott Sumner Scott Sumner
    1. January 2012 at 20:40

    123, I don’t agree about the base. It grew less than 1% from early May 2007 to early May 2008. So the problem was the supply of money, not the demand. Once we went into recession the level of interest rates fell, increasing the demand for money. At that point the problem was caused by the Fed not fully accommodating that increase in demand.

    But the initial problem was a low supply of money.

Leave a Reply