Interest rates versus the base

Here’s Paul Krugman:

But in this more complex world, where even the definition of the money supply becomes highly dubious, why even talk about an LM curve? Well, before 2008 most macroeconomists didn’t! They talked instead about interest rate targets, Taylor rules, and all that. Mike Woodford, who is probably our leading macroeconomist’s macroeconomist, has even made one of his signature modeling tricks the building of models in which there is (almost) no outside money. Sensible macroeconomists have known for a long time that quantity-theory type models, if they were ever useful, aren’t much use in the modern economy.

In normal times central bank monetary policy is conducted in terms of, and best thought of in terms of, the target interest rate

This is certainly the conventional view, but I think it’s wrong.  Let’s start with the fact that just as there are no atheists in a foxhole there are no non-monetarists during a hyperinflation.  When prices rise 8700% (almost) no one tries to explain the path of prices by referring to the path of interest rates. Even Wicksell and Keynes became quasi-monetarists during the early 1920s hyperinflations.  The reason is simple.  Interest rates tell us nothing about the level of prices and NGDP, whereas the base does.  Thus huge changes in the price level and NGDP can only be explained by looking at changes in the base.

[Matt Yglesias denies that money causes hyperinflation.  But all he's really saying is that the monetary deluge that causes hyperinflation has a REASON.  I.e. Latin American countries would choose to spend more than they received in taxes, and printed money to cover the deficit.  Countries with identical deficits, but good access to credit markets, would not print money and would not have hyperinflation.  It's not the deficit, it's the money printing.  As an analogy, Matt's claim would be like asserting that fiscal stimulus did not boost employment in 1942, WWII did.]

And if money explains hyperinflation, it also determines the path of prices and NGDP at lower growth rates, it’s just that the effect is disguised by the relatively greater importance of money demand (or velocity) fluctuations.  Conventional economists would claim that because velocity is volatile, interest rates are “more useful” way to think about monetary policy.  But they are not.

In the standard model, fluctuations in NGDP are caused by movements in interest rates.  And yet rates tend to be high during booms and low during recessions.  So how is the interest rate approach “more useful?”  Here’s where the NKs get clever; it’s not the interest rate that matters, it’s the market rate relative to the Wicksellian equilibrium rate.  Does this sound familiar?  Sort of like the monetary base relative to velocity.  Except that MV = PY is a tautology, whereas they merely have a theory.  At least unless you define the unobserved equilibrium rate as the one that produces steady growth in aggregate demand (PY).  In that case it’s also a tautology.  But how is it “more useful?”

Another problem with the nominal interest rate is that the policy lever locks up at zero rates, and hence central banks have trouble communicating at the zero bound.  In contrast, the base has no zero bound, and hence there is no point at which central banks are unable to communicate by adjusting a policy instrument.  For example, suppose the Fed indicated that they would keep increasing the size of their monthly QE purchases by 20% each month until expected NGDP growth got back on target.  That’s a clear strategy.  It would work very quickly (or else they’d own the entire universe quite quickly).  But a promise of low interest rates until some objective is met is basically consistent with Japan’s performance over the past 15 years.  You might do better, but you might not.

Another argument used in favor of the interest rate approach is that, out in the real world, people and policymakers think in terms of interest rates, not the base.  But that’s exactly the problem.  People think money has been easy since 2008 (even though according to Ben Bernanke’s criterion it’s been the tightest since Herbert Hoover was President), precisely because they’ve been taught that monetary policy is “best thought of” in terms of interest rates. They’ve been taught to look at not just a poor indicator, but one that is actually negatively correlated with the actual stance of monetary policy.  A poorly informed public will make bad public policy decisions.  And not just the public, even economists are confused.

At the same time the supply of base money is also an unreliable indicator (although less so than interest rates.)  Thus we should also not view changes in base money as a good indicator of the stance of monetary policy.  Rather changes in the base supply relative to base demand are what is important. As Ben Bernanke said, NGDP growth is the best indicator of whether money is easy or tight.

PS. OK, Bernanke said NGDP growth and inflation.  But he had to say inflation, NGDP was his choice.


Tags:

 
 
 

46 Responses to “Interest rates versus the base”

  1. Gravatar of nickik nickik
    30. August 2013 at 08:08

    It seams to me that Keynsians just want to somehow save there intrest rate models even if they know that the are less useful. Because if they admit the montary growth and fall (and thus resulting growth and fall in NGDP) is better then the basiclly have to admit that they are closer to monetarist then to actually old keynsiansm.

    I mean what is left of keynes in NK? Specially does who do not belive in the liquidity traps.

    So I kind of think of them a little bit like marxists, in that they change theory all the time but they never want to signal to outsider that they are not really marxists or keynsians any longer.

    So even if all market montarist ideas are exepted, the greater part of macro economist will not call themself market montarist, the will call themself Post-New-Keynsian or Modern Keynsians or Modern-Keynsian-Montary School.

  2. Gravatar of jknarr jknarr
    30. August 2013 at 08:26

    Interest rate management is mostly a post hoc reaction to changes in Wicksell’s natural rate equilibrium where demand for money is unclear, and supply is fairly clear — and policymakers shift rates in response to imputed changes in demand, with small changes in base money sufficient at higher interest rates, and huge changes in base money necessary at low interest rates.

    One of the more interesting hyperinflation observations is: there is never sufficient money on hand to keep up with prices. Ironically, hyperinflation economies are constantly starved of physical cash (hence the printing): meaning that there is overwhelming demand for currency. This supports the role of expectations. Money production in hyperinflation is a demand/expectation symptom, not a cause.

    Hyperinflation is a result of future expectations melting down, and (future-promise) financial assets being mass liquidated for present cash: huge demand for the intermediary currency, in other words.

    Unfortunately for us, this tends to happen at the zero bound, where demand for yield-bearing-near-cash-equivalents has reached its end point; and base money demand goes infinite.

    And note that much of the hyperinflated “money” has already been produced in reserves, but not yet been demanded into currency — yet a long stint at zero rates tends to boost currency demand, by weakening bank balance sheet carry (harming bank credit quality) — and producing zero opportunity cost to physical currency.

    The then-created currency is a hot potato — but was necessary to intermediate the exit from financial assets and the acquisition of real (inflation-protected) goods. Cash is printed to keep up with this liquidation demand — but the table was already set long before by financial asset appreciation — it just needs a future expectation reversal.

    Again, Banks- and Feds- (and their handmaiden industries) have likely promoted the idea that low interest rates signify policy easing, because you get to go begging to the bank to refinance, add debt, free up cash flow, and the US gets to sell a ton of debt and maintain a strong global reserve USD status — all via tight money. Good for them, not so much for you.

    The citizenry prefers greater income instead of all these refi- and EBT- opportunities that tight money policy has on offer.

  3. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    30. August 2013 at 08:28

    ‘Latin American countries would choose to spend more than they received in taxes, and printed money to cover the deficit.’

    That also appears to have been what Mugabe’s Zimbabwe did. Until they ran out of space to add zeroes to the Zim-$. Then they did the opposite, they stopped–really stopped–printing money. Guess what also stopped.

  4. Gravatar of Roger Sparks Roger Sparks
    30. August 2013 at 09:02

    We need more skill in measuring money supply changes.

    We seem to have agreement that government deficits increase the money supply, so an annual deficit would be a measured change. The cumulative deficits added over the years would be an accumulated money supply.

    We also seem to have agreement that bank loans are an increase in money supply, so the annual change in total loans outstanding would be a change in money supply. Unlike government deficits, the cumulative changes in loans are not accumulated money supply but, instead, are accumulated CLAIMS on money supply. This is the source of eventual crashes resulting from excessive borrowing.

    GDP not only measures economic activity, it also measures CHANGE in money supply. This is true because, on the average, every transaction is taxed which changes the money supply. The rate is easily calculated as Federal Expenses less Federal Receipts, both divided by GDP.

    As you suggest, we need to think of money supply in a somewhat different way. The change in money supply is much more important than base money in the short term, but base money supply balanced with accumulated loan obligations is the key to longer term stability.

  5. Gravatar of ssumner ssumner
    30. August 2013 at 09:20

    jknarr, I’m afraid that’s wrong. Real money demand plummets during hyperinflation, often by more than 90%. The growth in the money supply causes the hyperinflation, and as the opportunity cost of holding cash rises, money demand falls.

    Roger, That reminds me of answers I used to read when I taught EC101. There may be something there, but I have no idea what you are talking about.

  6. Gravatar of Arthur Arthur
    30. August 2013 at 09:42

    Actually there is a even stronger lower bound for the base. The monetary authority cannot in any circumstances supply a negative quantity of base money. Not very relevant thou.

  7. Gravatar of Franky Franky
    30. August 2013 at 10:18

    Scott,
    This whole discussion is a matter of interpreting the same model in different ways. That is why Krugman says “best thought of in terms of”.
    But here’s why I think interest rates are more helpful.
    You say monetary policy works through the hot potato effect: the money supply increases, people don’t want to hold the extra money so they exchange it for something else, but since in the aggregate they can’t get rid of it, overall spending goes up and NGDP goes up.
    But what about applying your argument in reverse? When the central contracts the money supply it increases the supply of bonds. So people who don’t want to hold the extra bonds exchange them for something else, but in the aggregate they can’t get rid of the bonds, so spending and NGDP go up.
    Why doesn’t this hold? What’s special about money? Here’s where interest rates come in. If the Central Bank just exchanges one asset for another, with no effect on any prices, there is no reason consumers and firms would change their spending decisions. And that’s what would happen according to modern asset pricing theory if the CB exchanges assets valued only for their expected future cash flows. But money is also valued for its use in transactions, and if that utility declines with the amount of money in circulation, then the quantity of money determines the interest rate that money holders are willing to forgo in order to hold money instead of other assets. Note that through arbitrage this affects the returns on all assets, not just risk free short term bonds. The reason people focus on the latter is because that rate reflects only this liquidity premium of money, and not other factors determining asset prices like risk premia. But all rates of return move when the short rate moves, and that’s what changes spending: a fall in rates of return makes saving less attractive and investment more attractive, so spending goes up and NGDP goes up.
    I think this is why people find it more helpful to think in terms of interest rates.

  8. Gravatar of TallDave TallDave
    30. August 2013 at 10:40

    This has quickly become my favorite MF quote:

    “A fourth effect, when and if it becomes operative, will go even farther, and definitely mean that a higher rate of monetary expansion will correspond to a higher, not lower, level of interest rates than would otherwise have prevailed. Let the higher rate of monetary growth produce rising prices, and let the public come to expect that prices will continue to rise. Borrowers will then be willing to pay and lenders will then demand higher interest rates-as Irving Fisher pointed out decades ago. This price expectation effect is slow to develop and also slow to disappear. Fisher estimated that it took several decades for a full ad- justment and more recent work is consistent with his estimates.”

    That’s roughly the inverse of what they’ve been doing since 1980, of course, which looks more like this…

    “Paradoxically, the monetary authority could assure low nominal rates of interest-but to do so it would have to start out in what seems like the opposite direction, by engaging in a deflationary monetary policy.”

    So clearly interest rates aren’t a good indicator even in normal times — even the delta in interest rates isn’t all that indicative, as Mark Sadowksi showed the other day.

  9. Gravatar of Petar Petar
    30. August 2013 at 10:43

    a very nice promise by BoJ deputy governor:
    Bank aims to achieve the 2 percent “price stability
    target” and will continue with the QQE as long as it is necessary for maintaining that target
    in a stable manner (Slide 5). The first feature I mentioned earlier is a commitment to definitely achieving the target, while the fifth is a commitment regarding the continuation of
    the QQE, such that the Bank will continue with the QQE for as long as it is necessary to achieve the target

    from comment section in this M.Nunes’ post: http://thefaintofheart.wordpress.com/2013/08/30/meanwhile-in-japan/#comments

  10. Gravatar of jknarr jknarr
    30. August 2013 at 10:46

    Scott, we may be talking about different stages of hyperinflation — I’m talking impetus and beginning, you may be describing destination (Cagan?).

    It’s like how a equity market selloff strengthens the USD temporarily — as stockholders liquidate, they demand cash as an intermediary, and the USD strengthens (and the Fed steps in as cash-provider of last resort if liquidation pressures are too great).

    In short, severe liquidation of assets creates base money demand, which prompts policymakers to create base money — i.e. the Lehman crisis, QED.

    If the debt liquidation expectation is seen as near-permanent (say driven by a secular reversal toward violently higher bond yields, credit crisis, huge future nominal growth, or FX declines), then the liquidation-forced-central-bank-new-cash-production creates more selling pressures in and of itself, i.e. more liquidation-demand base money production itself worsens future expectations.

    The end result, of course, is the collapse of real money demand — but only after the liquidation bid for cash is done, e.g. 90% the debt in the economy has been liquidated in numirare/real terms, and the base money has been duely produced.

    Bottom line, first the debt formation, then the savings expectation reversal, then the liquidation, then the base money production — which further viciously aggravates debt expectations.

    As debt liquidation demand tapers off, real money demand is collapsed (this is driven by financial asset liquidation and ensuing cash production — in other words, all the pent-up future demand from financial asset savings is dumped into current goods market.)

    The end (and organic purpose of) hyperinflation is the extinguishing of unsustainable liabilities/savings (and these liabilities are another guy’s savings) to bring balance between the potential “current good” (consumption) and potential “future good” (savings) market prices.

    We now have a global savings glut and global near-zero-rates (which is a liability glut on the flip side — and was created by tight money). This, one day, will face liquidation and potentially-hyperinflationary extinction when these expectations that savings will continue to provide future real consumption finally reverse, and pent-up savings demand is dumped into current prices.

    Note that we’ve never really seen a return from below-potential-NGDP with a fiat non-gold currency before (the last return from below-trend was after the Great Depression, when gold rooted real bond returns).

    Effective monetary stimulus in our pure-fiat-and-huge dollar bond market, alongside sizable idle capacity, might just trigger this hyperinflationary debt liquidation.

    The “Great Stagnation” or hyperinflation: it’s a policy choice between the old and the young.

  11. Gravatar of Max Max
    30. August 2013 at 11:10

    “For example, suppose the Fed indicated that they would keep increasing the size of their monthly QE purchases by 20% each month until expected NGDP growth got back on target. That’s a clear strategy. It would work very quickly (or else they’d own the entire universe quite quickly). But a promise of low interest rates until some objective is met is basically consistent with Japan’s performance over the past 15 years.”

    The first method is more credible if the purchases would be costly to reverse, but is a lack of credibility the problem? Wouldn’t you say the problem is with the objective?

  12. Gravatar of Max Max
    30. August 2013 at 11:18

    Patrick, “That also appears to have been what Mugabe’s Zimbabwe did. Until they ran out of space to add zeroes to the Zim-$. Then they did the opposite, they stopped–really stopped–printing money. Guess what also stopped.”

    The inflation stopped when people stopped quoting prices in Z$. It doesn’t do any good to print money you can’t spend.

  13. Gravatar of jknarr jknarr
    30. August 2013 at 11:36

    As an aside, this debt-liquidation-for-base-money argument shows the immense need for NGDPLT on both sides of the coin: one day, we may desperately need the upside discipline of LT.

    NGDP has been too long suppressed, and the debt market has grown too large in comparison (now 3.5x) as a result.

    Really, if savers feel that they will be able consume 3.5x the current US GDP in future real terms (alongside and competing with social security, medicare, military liabilities), all the more power to them.

    One day, markets may recognize that existing liabilities/assets will not, in fact, provide real future consumption — let alone a real rate of return — and bonds will be sold hard, and base money created to deal with the liquidation demand.

    Credible NGDPLT is very much needed to deal with the savings overhang, just as it is needed to deal with the NGDP undershoot.

  14. Gravatar of benjamin cole benjamin cole
    30. August 2013 at 12:08

    Superb blogging. Yes, taper up until you get results.

    I think QE works not only on expectations but because bond sellers have to put their money somewhere.

  15. Gravatar of Rob Rawlings Rob Rawlings
    30. August 2013 at 13:47

    “For example, suppose the Fed indicated that they would keep increasing the size of their monthly QE purchases by 20% each month until expected NGDP growth got back on target. That’s a clear strategy. It would work very quickly (or else they’d own the entire universe quite quickly).”

    If the Keynesian are right and the owners of the assets bought by QE programs are indifferent between them and money then quite a huge amount of assets may need to be purchased to move NGDP even a small amount. All that money held in reserves may not matter in theory but it does make people nervous just when we need them to be optimistic.

    You give the example of Latin American countries getting inflation just by financing deficits with new money. Isn’t that in fact route 1 to increasing NGDP quickly ? Finance tax cuts or (even better) subsidize final sales with new money and NGDP will be where we need to be in weeks not months.

  16. Gravatar of Garrett M Garrett M
    30. August 2013 at 13:58

    Franky said:

    “You say monetary policy works through the hot potato effect: the money supply increases, people don’t want to hold the extra money so they exchange it for something else, but since in the aggregate they can’t get rid of it, overall spending goes up and NGDP goes up.
    But what about applying your argument in reverse? When the central contracts the money supply it increases the supply of bonds. So people who don’t want to hold the extra bonds exchange them for something else, but in the aggregate they can’t get rid of the bonds, so spending and NGDP go up.”

    Actually, the price of the bonds would change to bring supply and demand into equilibrium. The difference with base money is that it’s price is nominally fixed, so when supply and demand change the price of all other goods and services must adjust. Wages are sticky, so the labor market adjusts more slowly than the goods market, hence AD shocks have real effects in the short term.

  17. Gravatar of dtoh dtoh
    30. August 2013 at 14:15

    A few things seem obvious.

    1. Anyone reasoning from nominal interest rates is a moron.

    2. Tautologies (e.g. MV=PY) are not particularly useful for figuring out monetary policy.

    3. Any attempt to correlate MB with PY is Quixotian unless you first subtract ER from MB.

    Scott can we agree to the following.

    4. The Fed can increase PY by buying sufficient financial assets (i.e. doing sufficient OMP).

    5. If real prices of financial assets rise (and expectations are unchanged) then there will be a marginal increase in the exchange of financial assets for real goods and services.

  18. Gravatar of TravisV TravisV
    30. August 2013 at 14:43

    Fun paragraph written by Ashok Rao:

    http://ashokarao.com/2013/08/30/what-would-a-wonks-perfect-policy-platform-look-like

    Get rid of the Department of Education and allocate every child into school by a random lottery. Public education is a bit (but not really) like the individual mandate. It works well if everyone uses it without segregation. There are big externalities in moving a rich kid from his bubble of a rich school to a poorer school because support from his parents will make everyone in the poorer school better of. For free! If you think about “parental positive influence” as a scarce good concentrated in the top 20% of the population, there is huge, huge inefficiency in having many rich kids go to the same school. In this case, redundancy is bad.

    http://ashokarao.com/2013/07/22/radically-centrist-education-a-thought-experiment

  19. Gravatar of TravisV TravisV
    30. August 2013 at 14:44

    I liked these two sentences written by Ryan Avent:

    “For all I know the president is a die-hard Scott Sumner fan and Mr Summers has whispered to him that market monetarism is now his macroeconomic lodestar. But there are few signs that anything like that is going on.”

    http://www.economist.com/blogs/freeexchange/2013/08/monetary-policy-1

  20. Gravatar of Franky Franky
    30. August 2013 at 14:50

    Garret M, it is precisely when the CB changes the money supply that the price of bonds, i.e. the interest rate, moves. Because that is the cost of holding money. If the CB is just exchanging, say, short term for long term bonds, then bond prices will not move. Holding money constant, bond prices are determined by the present value of their cash flows, not supply and demand.

  21. Gravatar of Bill Ellis Bill Ellis
    30. August 2013 at 16:54

    Have you guys seen this yet?

    How Stimulatory Are Large-Scale Asset Purchases?

    Here’s the conclusion to a FRBSF Economic Letter from Vasco Cúrdia and Andrea Ferrero on the question of How Stimulatory Are Large-Scale Asset Purchases?:

    … Asset purchase programs like QE2 appear to have, at best, moderate effects on economic growth and inflation. Research suggests that the key reason these effects are limited is that bond market segmentation is small. Moreover, the magnitude of LSAP effects depends greatly on expectations for interest rate policy, but those effects are weaker and more uncertain than conventional interest rate policy. This suggests that communication about the beginning of federal funds rate increases will have stronger effects than guidance about the end of asset purchases.

    http://economistsview.typepad.com/economistsview/2013/08/how-stimulatory-are-large-scale-asset-purchases.html

  22. Gravatar of Jason Jason
    30. August 2013 at 17:02

    The way I see it, interest rates the monetary base and NGDP are all inter-related …

    log r ~ log NGDP/MB – log k

    Plotted here:

    http://informationtransfereconomics.blogspot.com/2013/08/the-interest-rate-in-information.html

    [I think one of the major reasons I started a blog is so I could post pictures in Scott's comments.]

  23. Gravatar of Jehu Jehu
    30. August 2013 at 17:16

    Why not just buy out all the other currencies, i.e., flood the world market with dollars? Bernanke has already explained this should work in 2002. They have to go in any case. :)

  24. Gravatar of Philo Philo
    30. August 2013 at 18:31

    “[I]nterest rates . . . [are] actually negatively correlated with the actual stance of monetary policy.” This statement would be fine if ‘stance of monetary policy’ meant what *you* mean by it, namely that it is tight if it is tighter than it ought to be and loose if it is looser than it ought to be. But you encounter a (rhetorical, communicative) difficulty, in that most people don’t mean this by ‘stance of monetary policy’. For them, ‘tight’ means *tighter than X* and ‘loose’ means *looser than X*, and X ≠ *what the policy ought to be*.

    I hope this comment is helpful. I would be more helpful if I could tell what X *is*, but unfortunately I haven’t been able to figure that out. You might ask around among people who use the terms ‘tight’ and ‘loose’ [‘easy’, ‘accommodative’], but I suspect that most of them are too confused to give you a clear answer.

    Maybe you should first try to convince people to adopt your definition; that might make the rest of your task easier.

  25. Gravatar of Philo Philo
    30. August 2013 at 18:36

    @ Benjamin Cole:
    “Yes, taper up until you get results.” But evidently the FOMC is *already* getting the results it wants; that’s why it’s preparing to taper *down*. There’s not much point in advising them how to hit your target, when they’re actually aiming for something else. (I’m not sure what!)

  26. Gravatar of ssumner ssumner
    30. August 2013 at 19:11

    Franky, Money is more important than bonds because money is the medium of account and bonds are not. It’s that simple. Regarding transmission mechanisms, I agree that asset prices matter (but that’s different from saying interest rates matter.) But you ignore expected future changes in NGDP, which are all important—and driven by the hot potato effect.

    jknarr, Increased money demand is deflationary. And hyperinflation doesn’t come from debt bubbles.

    Max, Central banks trying to inflate have NEVER had a credibility problem. NEVER.

    Rob, Under NGDPLT very little money creation would be needed, far less than what we’ve already done.

    dtoh, I agree on all five points.

    Travis, I got that in a blog post already, but it’s not yet posted–huge backlog.

    Bill, That paragraph is a mess, I suppose I’ll have to do a post.

    Jason. Yup.

    Jehu. That would “work” but much less than that is needed.

    Philo, I mean what Ben Bernanke means. Tight money is slow NGDP growth and easy money is fast NGDP growth. Yes, the thinking of most people is hopelessly muddled.

  27. Gravatar of Franky Franky
    30. August 2013 at 20:37

    Scott, how does money being the medium of account change people’s behavior?
    Let’s take an open market operation. The CB buys bonds in exchange for money. People aren’t wealthier, they just have a different asset portfolio, with more money and less bonds. So there’s no reason they will consume or invest more. They may want to exchange money for other assets, but since they can’t do that in the aggregate, interest rates will fall by enough so that people are willing to hold the extra money. If you ignore the effects of interest rate changes on spending, what else happens?

  28. Gravatar of James in London James in London
    30. August 2013 at 22:25

    Jehu. See Recent monetary policy in Switzerland. The world was so desperate for Swiss francs it was driving the price through the roof, so they started printing new ones. That soon stopped that!

  29. Gravatar of ssumner ssumner
    31. August 2013 at 06:16

    Franky, There are lots of flaws in that comment, and I suggest you take my short course on money in the right margin. The MOA role of money is all important. When other goods markets are affected, we don’t ask for a change in wealth in order to explain a change in value. Do apple prices fall after a big harvest because of a change in wealth? Obviously not. How about gold? Now make gold money, then what happens after a big gold discovery? How is paper money different?

    As far as transmission mechanisms, the most important is the expected future change in NGDP due to the HPE. Woodford also emphasizes this channel, although not via the HPE. The second biggest is changes in asset prices; stocks, risky bonds, commodities, real estate, foreign exchange, etc. Short term risk free rates have an effect, but it is minor in comparison. People don’t go out and buy a new home because 3 month T-bill yields fall.

    James, Good example.

  30. Gravatar of Franky Franky
    31. August 2013 at 07:11

    Scott, I’m really not following you here. All I am saying is that an increase in the money supply changes behavior through a change in the price of holding money, i.e. interest rates. I’m not saying you need a change in wealth to explain anything, just that if you ignore the interest rate, and since there is no change in wealth, I see no other effect on behavior (ignoring the new monetarist effect of carrying around a larger real money balance).

  31. Gravatar of Franky Franky
    31. August 2013 at 08:14

    Let me elaborate a bit.
    I understand the intuition of saying: money is the unit of account, therefore if you increase money, the prices of everything else must go up in proportion. But you need to think about how exactly this happens.
    Take a consumer optimizing his lifetime utility. Each period he consumes a fraction of his expected lifetime wealth, a fraction that depends on the discount rate, the real interest rate and the elasticity of intertemporal substitution. This is just the standard Euler equation. Do you buy this or do you have a different model in mind? (if you do it would be really helpful for those of us struggling to understand you if you could write down an equation describing consumer behavior)
    Now the CB buys some bonds from him in exchange for money. How does this change his optimal consumption plan? He is not wealthier, so he does not mechanically spend more if he just sticks to the fraction of lifetime wealth he was planning to consume. There has to be another reason making him go out and spend a larger fraction of that wealth. You may say: the reason is that consumers are holding more money than they want to, so they will go out and spend it on consumption goods. But that’s not optimal in this model. They are better off exchanging their money for bonds, pocketing the interest, and sticking to their original plan. Of course, they can’t do this in aggregate and so the interest rate on bonds will fall until consumers are willing to hold the extra money. But ignoring the effects of this change in interest rates, they will hold on to the money, not consume more. So something needs to trigger the additional spending.
    In the NK model that something is the fall in the real interest rate. It makes consumption today relatively cheaper than consumption in the future. So the consumer spends a larger fraction of his wealth today.
    Can you say exactly at which step you disagree, and/or provide your own model?

  32. Gravatar of Franky Franky
    31. August 2013 at 09:22

    One more post to try to zero in on where I think the disagreement is. I just went through your course on money and the key point is:

    “When the Fed increases the supply of base money, people try to get rid of excess cash balances. Individually they can do so, but collectively they cannot. The paradox is resolved by the fact that when people try to get rid of excess cash balances, prices rise until the public wants to hold those extra cash balances”

    What I am saying is that prices of consumption goods don’t just go up by magic, people need to spend more on consumption and then firms will raise prices in response. So I think our disagreement lies here: you claim people will try to get rid of excess cash by spending more on consumption, and I am saying they will do it by buying bonds. This is crucial, because it determines which prices adjust DIRECTLY, consumption prices or interest rates. If consumption prices adjust directly then you are right, interest rates may not matter much for transmission.
    But that is not optimal behavior in the standard model of consumer behavior that I described in the previous post. In that model, the consumer tries to rebalance his portfolio to bonds, interest rates fall, and then consumption rises because of the fall in interest rates. Finally, firms raise prices in response to the additional spending by consumers.
    Is this is where we disagree, and do you have a different model for consumer behavior?

  33. Gravatar of Petar Petar
    31. August 2013 at 09:33

    One Carney thinks its about what Fed buys:
    What makes this circular and annoying is that demand for base money is simply a Higgs boson particle: something that we can’t directly observe but we assume is there because without it nothing else we think we know makes much sense. So Sumner knows that base demand must be increasing if growth is below target while base supply remains steady or increases because that’s the thing that would explain below target growth.

    Sumner thinks the Fed could “loosen” policy by announcing that it would grow its quantitative easing program by 20 percent each month until growth returned to its target. I think that’s right—but only if the Fed made a dramatic change in what it buys.

    http://www.cnbc.com/id/101000307#_gus

  34. Gravatar of Max Max
    31. August 2013 at 10:05

    “Central banks trying to inflate have NEVER had a credibility problem. NEVER.”

    Ok. Then I don’t understand why a promise of low interest rates until the objective is met wouldn’t work just as well as increasing the size of their monthly QE purchases by 20% each month until the objective is met. What’s the difference, if both are equally credible?

  35. Gravatar of DOB DOB
    31. August 2013 at 18:17

    @Max, you’re exactly right.

  36. Gravatar of Tom Brown Tom Brown
    31. August 2013 at 18:25

    “…they merely have a theory.”

    The use of the word “theory” in economics is apparently different than in science. The theory of quantum mechanics is “merely a theory” …but I guess you’re right! A tautology “trumps” theories. Tomorrow QM could be in the trash, but A = A+0 will still hold.

  37. Gravatar of Tom Brown Tom Brown
    31. August 2013 at 18:52

    Scott, the other day you wrote that the HPE is “very weak” with “zero rates.” But if the CB were to actually follow through and “keep increasing the size of their monthly QE purchases by 20% each month until expected NGDP growth got back on target” that would no longer be a weak HPE, right?

    I get that it’s all in the communication, and that this 20% thing is probably not necessary to actually carry out… at least for long. And I’ve heard you say that “threats” aren’t necessary, but that sounds like a threat!

    Are you saying that the expected HPE (due to this credible …er “promise” from the Fed) is equivalent to:

    “changes in the base supply relative to base demand are what is important?”

    … or could we even say

    “expected changes in the base supply relative to base demand are what is important?”

    And is it really the base that’s so important? Those expected asset purchases will swell reserves, but also (more importantly perhaps?) put bank deposits (MOE, as Nick Rowe would say… *I think*) into the hands of the non-bank public sector… sure, those are deposits they could swap for base money (physical cash) …but I presume this scheme would work in a cashless society too(?), so it’s hard to believe this form of base money (cash) is vitally important here.

    Just for laughs say we were cashless… do you still make these statements using the term “base money” (now Fed deposits only)?

  38. Gravatar of ssumner ssumner
    1. September 2013 at 08:05

    Franky, You said;

    “Scott, I’m really not following you here. All I am saying is that an increase in the money supply changes behavior through a change in the price of holding money, i.e. interest rates. I’m not saying you need a change in wealth to explain anything, just that if you ignore the interest rate, and since there is no change in wealth, I see no other effect on behavior (ignoring the new monetarist effect of carrying around a larger real money balance).”

    And I’m saying this is completely wrong, for all sorts of reasons. Some monetary shocks are huge, but have no impact on interest rates–say the 1933 dollar devaluation. And even when short term risk free rates change, they have very little effect on NGDP, the main effect comes from other mechanisms.

    In your longer paragraph you are trying to explain why people would buy more real goods, whereas I am claiming they buy more nominal goods. The real effects in my model come from stickiness. You are simply assuming stickiness and going right for the real effects, I’m saying that’s wrong, that you need to first explain the nominal effects (via the MOA) and then figure out the real effects that result from the nominal shocks. A change in the MOA changes NGDP, I think we all agree on that. Then if wages are sticky more NGDP leads to more employment and output. That’s part is completely beyond dispute. Once it is produced, it becomes part of GDP, even if inventory. Real consumer behavior is the tail, not the dog. QED.

    But if you insist on a consumer model, just have them expect the previous events to happen. That means more real output, hence more real income, hence more real consumption.

    Now of course you can insist that in my model the interest rates is falling relative to it’s unobserved Wicksellian rate. And that’s true. But the proof’s in the pudding. Actual real world Keynesians think and talk in terms of actual interest rates, not spreads between market rates and unobserved Wicksellian equilibrium rates. Just read my other recent posts.

    Max, Because the first promise is consistent with it taking a million years. What good does it do to be credible, if you have no time frame!

    Petar, It would not need to change what it buys, because it would not need to buy anything.

    Tom Brown, I hope you are not the Tom Brown who left this comment at Cullen’s blog:

    “Sumner does seem to have learned something. It’s not clear what, but in his post today he was talking about how “base money” is “endogenous” in an “ideal world.”
    That’s REALLY a change from three years ago!”

    As if you are, it would suggest you are in way over your head here. Sorry to be so blunt. My views haven’t changed one iota.

    Yes, any form of base money allows for a nominal anchor, and you completely misunderstood my tautology joke.

  39. Gravatar of Tom Brown Tom Brown
    1. September 2013 at 09:07

    Scott,

    “As if you are, it would suggest you are in way over your head here. Sorry to be so blunt. My views haven’t changed one iota.”

    That was definitely me. Sorry If I’ve misinterpreted you. It was my impression that they had changed (I’m specifically referring to threads I saw you engaging in with Scott Fullwiler on another blog from perhaps 3 years back). I’m much more interested in learning what your views are now whether they’ve changed or not, and I’ll take you at your word that they haven’t. Sometimes I get frustrated because it appears to me there are contradictions, but I’m happy to admit I’m wrong about that. My views very much are changing… all the time as I learn more in general. So my apologies If I got it wrong.

    Your tautology joke?… you mean my response to W. Peden on the other post? Yes, that was a stupid blunder on my part (due to me misreading W. Peden) … but I’d already apologized for that, but you mean there’s more I misunderstood? Good grief!

  40. Gravatar of Tom Brown Tom Brown
    1. September 2013 at 09:21

    Scott,

    Let me ask you this, do you agree with this?:

    “Tom: thanks, but that’s not quite right. Under inflation targeting the quantity of money is endogenous in both the short run and the long run. It’s the nominal rate of interest that is exogenous in the very short run (6 weeks or less, for the central bank anyway), but endogenous in the long run.”

  41. Gravatar of ssumner ssumner
    1. September 2013 at 11:17

    Tom, I just meant my tautology /theory distinction looks stupid unless you include the rest of the paragraph, which was meant to be funny.

    Yes, I’ve always argued that the base can be viewed as exogenous in some contexts and endogenous in others. It depends what question is being asked. If someone says point blank it is endogenous, and hence non-controllable, I disagree. It’s only endogenous in the context of some other target. In a technical sense it is exogenous, in the sense that the Fed has the technical ability to move it around. By doing so they affect other variables. If they target another variable then it becomes endogenous, as it must be adjusted to keep the other variable on target. Read Nick Rowe on endogeniety.

    Yes, I still agree with that quote.

  42. Gravatar of Franky Franky
    1. September 2013 at 14:45

    Scott,
    You said: “In your longer paragraph you are trying to explain why people would buy more real goods, whereas I am claiming they buy more nominal goods.”

    I am not, I am talking about nominal goods and nominal spending. The only reference I made to real effects is in the last paragraph about the NK model and the fall in the real interest rate, but that was not necessary. Just replace NK with RBC and real with nominal and my description still holds: nominal spending goes up when the nominal rate falls. Then prices rise, bringing real money balances down which in turns brings the nominal interest rate back up. All of this happens instantaneously in a flexible price model, but those are the forces at work.

    “A change in the MOA changes NGDP, I think we all agree on that.”

    A change in money changes NGDP, yes, the question is how. It seems to me that you are still not being clear about that. You assume that it follows from money being the MOA, but that by itself does not change behavior. Consider the following economy:

    There are two types of goods, oranges and apples. And there are two kinds of apples, red apples and green apples, which are perfect substitutes in consumption. The only difference is that red apples are the MOA.
    The central bank buys some green apples from consumers exchanging them for red apples, increasing the supply of red apples but keeping total apples the same. Does the price of oranges (in terms of red apples) go up? No, because consumers want to eat the same ratio of oranges to (total) apples, and the quantity of apples and oranges hasn’t changed. Red apples being the MOA changes nothing by itself.
    You need some other factor driving people to spend more on oranges to change their price. In the case of money, it is that money is the MOE and therefore not a perfect substitute for bonds.

  43. Gravatar of Max Max
    1. September 2013 at 17:11

    “Because the first promise is consistent with it taking a million years. What good does it do to be credible, if you have no time frame!”

    If expectations don’t instantly change, it can only be because people don’t believe the central bank will stick to the policy. In other words, it’s not fully credible. In that case, it helps to do something that ties its hands, something that is costly to reverse.

  44. Gravatar of DOB DOB
    2. September 2013 at 10:41

    Hi Scott,

    I have a comment awaiting moderation on this thread (I believe that’s because I linked to JPKoning’s blog)

    Thanks,
    DOB-

  45. Gravatar of DOB DOB
    3. September 2013 at 08:06

    [Trying again as suggested. Removing link to JPK's blog to avoid filter]

    Scott,

    I’m responding here to your response to my comment because this conversation really belongs under this post.

    You wrote: “I agree that a swap of $20 bills for $100 bills does nothing, as they are perfect substitutes. But T-bills are not a MOA, as their price can change, even if only a tiny amount.” and “If there is another financial asset that is identical, it is essentially cash, and should be counted as part of the base.”

    I have to strongly disagree here: the price of an overnight repo does not deviate from par, not even by a tiny amount. Should we count it as part of the base? Does that mean that when the Fed increases the base via repos (which it often does, and in my opinion that should be the only instrument the Fed ever uses) the base is actually unchanged (since the repos are “negative base”) ?? Does that mean that when it buys short term bonds which deviate from par ever so slightly, the base changes by say, 10% of the amount of the transaction? Or do you actually have a discontinuity in your model whereby if the instrument can vary in price–at all–it is considered completely “non-base”.

    These are the type of aberrations we get to when we think in terms of quantity.

    You also wrote: “1. If the monetary injection is expected to be temporary, then NGDP will be expected to show zero growth in 3 years.

    2. If enough of the injection is expected to be permanent enough in 3 years to raise the base by 15%, then the expected level of NGDP three years out will be 15% higher.

    3. If the entire monetary injection is expected to be permanent, then NGDP is expected to triple in 3 years.”

    I roughly agree with that, but that’s an incredibly convoluted way to think about things. So the Fed must move the base from 1 to 3 and the market is somehow supposed to infer from that that it will return it to 1.15? Why? Why not to 1? Why not to 1.5? Why not just move it straight to 1.15 and leave it there? What’s the point of going to 3?

    Another limitations of quantity-oriented thinking.

    I’ll concede that it is possible to reason in terms of quantity and generally get to the right conclusions. And in fact I think you generally do and I agree with most of what you say.

    But in the above post, you use the zero bound as an example for why we should think in terms of base rather than interest rates specifically because we’re at the zero bound when in fact quantity-oriented reasoning completely breaks down at the zero bound much in the same way that classical mechanics isn’t going to tell you much regarding the relative speed of two photons flying at each other.

    I guess I’ll compare quantity-oriented reasoning to classical mechanics: it works when nominal rates are away from the zero bound, with well behaved demand for liquidity / velocity.

    Interest rates-oriented reasoning is kind of like relativity in that it works in all cases. The physics analogy isn’t perfect as I find IR-reasoning much more intuitive than Qty-reasoning, so I prefer to use it even in cases were Qty-reasoning would produce the correct answer.

    IR-reasoning enables you to properly orthogonalize two features of money:

    - Medium of Exchange: we use money for transactions, to make payments, and need liquidity to carry out transaction. Do we need 3x more liquidity in a crisis than in normal times? Of course not! We carry out fewer transactions if anything.. The quantity of medium of exchange in existence is largely controlled by the spread between the lending rate and the deposit rate (Fedfunds and IOR if you will). JPKoning calls this non-pecuniary return on money or convenience yield.

    - Unit of Account: the price level. NGDP. And ultimately unemployment. That’s controlled by varying Fedfunds, which essentially moves the real return on money relative to the natural real rate (is that what you call the Wicksellian equilibrium rate?)

    Once you think in terms of these two levels: the lending rate and the deposit rate), everything becomes clear:

    - Pushing Fedfunds against IOR is associated with base soaring and doesn’t mean anything in terms of the price level.

    - The zero bound does indeed prevent immediate control of return on money though the Fed could commit on future nominal rates rather than make random increases in today’s base with some hope that the market interprets that in terms of future base.

    - Pegging IOR to 0% is unnecessary and causes swings in base that do not reflect the reality of transactions being carried out (i.e. velocity fluctuations). It also causes some people to funge medium of exchange and unit of account into “medium of account” :)

    - With IOR pegged at 0% or meaningfully lower than Fedfunds, the base could go down over the next 20 years without prices moving down and electronic means of payments mean we need to hold less and less money. As Krugman points out in the post you cite, Woodford pushed this to the extreme where qty = 0 at all times.

    - Sufficiently negative interest rates are the silver bullet when it comes to generating inflation. Commitment to keep rates at 0% until back on path is second best.

  46. Gravatar of TheMoneyIllusion » Yes, expansionary monetary policy really did cause the 1970s inflation. TheMoneyIllusion » Yes, expansionary monetary policy really did cause the 1970s inflation.
    6. September 2013 at 05:00

    [...] a reason; to prevent high unemployment.  In his view that made it a non-monetary phenomenon.  I recently criticized a Matt Yglesias post on the same grounds.  He had claimed hyperinflation was not a monetary [...]

Leave a Reply