Why are macroeconomists so obsessed with interest rates?
If I wrote a macro textbook, I would try to avoid any mention of interest rates or inflation. The Fisher equation would use expected NGDP growth. The AS/AD model would use hours worked as the real variable and NGDP as the nominal variable. The transmission mechanism would not involve changes in interest rates, but rather the excess cash balance mechanism. More cash would raise expected future NGDP. This would raise the current price of stocks, commodities and real estate. (As in Islamic economics, there’d be no interest rates.) The higher asset prices would tend to raise current AD. More nominal spending, when combined with sticky wages and prices, would boost output.
[Update 10/22/10: This is why I shouldn’t do 4 posts in one day. Commenters pointed out two flaws. The Fisher equation uses interest rates. And interest rates are important for inter-temporal decisions. How about this: A macroeconomics free of the concept of inflation, and a business cycle theory that did not use either inflation or interest rates. Is that slightly less crazy?]
But obviously I’m the exception. When rates hit zero and the Fed couldn’t move them anymore, I expected economists to shift over to some other mechanism; the money supply, CPI futures, exchange rates, etc. Instead they started talking about how the Fed could promote a recovery by lowering long term rates. (But if the policy is expected to work, wouldn’t it boost long term rates?) Or they talked about how the Fed could reduce real interest rates by boosting inflation. Some even argued that the Fed would have to boost inflation expectations to 6% in order to get a robust recovery, forgetting that this view directly conflicts with another key assumption of Keynesian macroeconomics—that the SRAS is very flat when unemployment is high.
Some of my commenters argued you couldn’t raise NGDP without first creating inflation expectations, which would lower the real rate of interest. But that’s not necessary at all. If you create higher NGDP growth expectations, then even if expected inflation doesn’t rise at all, the Wicksellian equilibrium real rate will rise and monetary policy will become more stimulative. So the Fed doesn’t need to lower real interest rates in order to stimulate the economy. Conversely, a policy such as interest on reserves might have had a devastating impact on aggregate demand, even if it had no impact on interest rates.
I’m genuinely confused. When we explain why a big crop of apples makes NGDP in apple terms rise sharply, we don’t resort to convoluted explanations involving an interest rate transmission mechanism. Why then, when there’s a big increase in the supply of money, do we think it will only affect nominal GDP in dollar terms via some sort of interest rate transmission mechanism?
Tags: New Keynesianism
21. October 2010 at 15:11
True. But many an academic reputation (Bernanke´s included) was made on “devising” new “channels”. In B´s case it was the “credit channel” of which an elementary exposition by the chairman himself is linked below:
http://www.philadelphiafed.org/research-and-data/publications/business-review/1988/brnd88bb.pdf
21. October 2010 at 16:50
Here’s a gross historical oversimplification and answer to the headline question:
1) Age of Keynes (1936-1963)
There’s only one MTM: interest rates. Interest rates are not very effective, ipso facto monetary policy is completely ineffective.
2) Age of Friedman (1963-1993)
A brief interlude between the two ages of interest rate insanity.
3) Age of Taylor (1993-?)
“Oh God, how come low interest rates aren’t working? We’re all going to die!!!!
21. October 2010 at 17:32
The idea behind the fed controlling interest rates is simple enough that the average reporter can understand it. So they get covered more than other types of macro interventions, which raises their salience in the rest of the population. Actually, I think this gells with what you were saying about stiglitz. He has no opinion about monetary policy because in his (and the popular) mind, monetary policy and the fed rate are synonymous terms. I’d actually love to see a list of all the things the fed is currently doing right now that might be considered monetary policy. I have no idea where to find such a things, but some of them, like paying interest on reserves, are clearly very important to the macro outlook.
21. October 2010 at 17:34
I have a tough time wrapping my head around it and therefore probably confuse my students because the textbook standard was all I was taught.
When I teach about NGDP targeting I always explain it as (PxY). So, if we’ve had some shock that causes Y to decrease by -5% and we have a 5% (PxY) growth target, the central bank needs to pursue inflation of 10%. When Rogoff is calling for 6% inflation target for an extended period, that’s what I hear him thinking–even though it might not be.
Students understand the Fisher equation, so the idea of creating higher inflation expectations to get negative rates and getting banks/corporations to stop hoarding money makes sense to them. When I hear a Fed president call for price level targeting, I hear “we need to raise inflation expectations to get people to stop hoarding cash.” If Rogoff isn’t thinking “higher NGDP growth” then I doubt any of the Fed presidents are either.
Yesterday I covered monetary aggregate targeting, inflation targeting, level targeting, and threw in the NGDP target. And every time I do it sounds like a free lunch.
I think tomorrow I will go back and talk about Friedman’s inflation expectations targeting (via Beckworth’s article today) and then do NGDP expectations targeting. But I won’t feel confident with either because I was taught neither.
21. October 2010 at 17:38
In the comment above I should say “if we are experiencing a shock to Y of -5% then the Fed should be pursuing 10% inflation.” I understand the forward-looking (expectations) part is what is important.
21. October 2010 at 17:44
It seems people forget the Federal Reserve is a central bank, including most of those who work there. Like any other bank, the Fed is a financial intermediary, i.e. it coordinates the plans of savers and borrowers by expanding and contracting the credit supply. When banks hold excess reserves (or individuals hold currency) they are saving by lending to the Fed, and it is the Fed’s obligation to invest those resources in some manner. The Fed doesn’t create new loans directly, but instead purchases assets. The former holders of those assets then spend their new money — probably creating new loans in the process.
A liquidity trap occurs when the former holders of assets the Fed purchases do not spend their new money. If the money is held, then the Fed has failed to invest the resources that were lent to it. If the Fed were legally permitted to buy only one kind of asset, and its holders would just as soon hold money instead, then this “liquidity trap” would be a really serious problem. The Fed would be constitutionally unable to fulfill its role as financial intermediary of last resort (so to speak). But fortunately, that is just not the world we live in, and so it just really doesn’t matter.
Is this a correct description of the situation? I am no economist, and so perhaps I have missed something. In any case, there is no need to mention interest rates — they are almost an epiphenomena.
21. October 2010 at 17:44
I’m sorry for a 3rd comment. I wanted to point out that I definitely don’t see the monetary transmission mechanism as simply lowering interest rates, as Stiglitz, Krugman, and apparently most economic forecasters do. I think that just goes back to their econometric equations. “An X TRILLION purchase of 30 year Treasury bonds will cause Y% decrease in yield and translate to Z% increase in investment and GDP growth.” Their equations don’t capture inflation expectations or NGDP growth expectations so they don’t fathom it.
Stiglitz and others don’t even seem to ever think of the Fisher equation…
21. October 2010 at 17:54
JTapp,
You wrote;
“Yesterday I covered monetary aggregate targeting, inflation targeting, level targeting, and threw in the NGDP target. And every time I do it sounds like a free lunch.
I think tomorrow I will go back and talk about Friedman’s inflation expectations targeting (via Beckworth’s article today) and then do NGDP expectations targeting. But I won’t feel confident with either because I was taught neither.”
I myself am teaching a new undergraduate course of my own creation on Monetary Policy. I feel very conflicted between what I myself was taught and I what I deeply feel is currently relevant. But each class I manage to throw in a “surprise” concerning recent monetary developments. And my semester project is for each of the students to craft a recommendation for the FOMC meeting in December.
While I’ve hinted at expected NGDP level targeting I haven’t yet addressed the topic head on. Eventually I’m going to have to screw up enough courage to reveal where our class is inevitably going.
21. October 2010 at 18:00
Mark, thanks for the encouragement. Hope your students are enjoying it. About 1/3 of mine are. :^)
21. October 2010 at 18:09
Marcus, It just seems to me that when people talk about monetary policy affecting the economy, all they talk about is the effect on interest rates. I regard interest rates as a side issue.
Mark, That about sums it up.
cassander, I recall Rogoff once criticized Stiglitz for arguing that high interest rates during the East Asian crisis were a sign of tight money. Many of these countries were suffering from very high inflation, and the high rates reflected the Fisher effect. That should have been obvious, but Stiglitz didn’t seem aware of that fact.
JTapp, I’m glad Beckworth publicized that aspect of Milton Friedman’s views on monetary policy. I’ve also done some posts on that, and it’s important that people are aware that the conservatives/monetarists of today do not represent Friedman’s views.
Lee Kelly, I don’t think of the Fed in terms of credit, I think of them in terms of cash. If they swapped cash for gold the effect would be roughly the same.
JTapp and Mark, I actually don’t include much of my blog in my class. I’ve never spent any serious time on NGDP targeting, or futures targeting. But I do teach what’s in Mishkin’s book–monetary policy is highly effective at zero rates, nominal rates don’t reveal the stance of policy, you have to look at other asset prices, etc.
21. October 2010 at 18:18
When we get a huge crop of apples, there is a yield effect down the apple curve. Apple inventory down the chain puts pressure on retailers to sell more apples for cheaper. Produce managers at grocers read the produce reports, they know the apple yield on the farms is way up. Apple pie bakers are reading that same report, The predicted yield effect is Apple Illusion. It is Apple Illusion because the effect takes place before price equilibrium, it is all based on supply side asymmetric knowledge.
21. October 2010 at 18:20
Scott: I think this is the right question to ask. But I think your discussion of the slope of the SRAS curve is a red herring. The question is: why should the transmission mechanism from money to the quantity of output demanded, at a given price level, (i.e. the rightward shift in the AD curve at a given P) be seen in terms of interest rates? That question is independent of whether the SRAS curve is horizontal or vertical.
I think Lee is onto something. Because the central bank is seen as a *bank*, that is seen as borrowing and lending, and buys and sells nominal assets (government bonds) it is seen that monetary policy “must” work through interest rates. The fact that the percentage change in the number of bonds in public hands is normally trivially small, compared to the percentage change in the stock of base money, should tell us there’s something wrong with this approach.
21. October 2010 at 18:25
JTapp,
The enthusiasm divide is very wide. Some are there solely for the credit. Some, in contrast, are incredibly interested in the subject. But mostly I notice they enjoy it much more when I say what’s truly on my mind.
Scott,
You wrote:
“JTapp and Mark, I actually don’t include much of my blog in my class. I’ve never spent any serious time on NGDP targeting, or futures targeting. But I do teach what’s in Mishkin’s book-monetary policy is highly effective at zero rates, nominal rates don’t reveal the stance of policy, you have to look at other asset prices, etc.”
That’s because of your modesty. I haven’t included economics blog entries in my class readings yet, although I’ve seriously thought about it (and you can be sure that they do influence what I teach). I agree that Mishkin is the “good book”. But ultimately I think I’m going to tread where you haven’t sufficient self effacement to go.
21. October 2010 at 18:40
Mark: I think that seeing monetary policy as operating through interest rates pre-dates Keynes. Yep. Hawtrey 1938: “A century of Bank Rate” http://www.jstor.org/pss/1825602 . OK, post-dates Keynes’ GT, but non-Keynesian, and writing of a long tradition.
And cassander: it’s not just that the interest rate story is simple enough for reporters to understand; this is how most monetary economists view the monetary transmission mechanism too.
Suppose central banks bought bricks in open market operations. Would we have a story of the transmission mechanism in which the price of bricks played the leading role?
21. October 2010 at 18:45
Nick Rowe,
I’m aware of that fact (as well as the tradition that predates Hawtrey). That’s why I qualified it as a “gross historical oversimplification”. But your comment is very useful to those who are not.
21. October 2010 at 18:54
Mark: Yes, sorry. I should have noted your “gross oversimplification”.
21. October 2010 at 18:58
“(But if the policy is expected to work, wouldn’t it boost long term rates?)”
Why does everyone struggle with the idea of higher long term rates?
If you flood the market with REALLY cheap hard assets, the demand for money / credit increases and when the supply isn’t endless – rates to borrow go up.
If only we had a HUGE pile of really cheap hard assets to sell off?!? Oh yes, we do have such a pile…. we just need to wipe out the insolvent banks.
Maybe this will help you to imagine it… say the FED was going to aggressively tighten monetary supply at the same time that the US government was going to dissolve and sell off ALL its assets in an auction 180 days from now – the national parks, the millions of acres, all the mineral rights, all of its buildings, everything.
Would demand for money / credit go up?
If the answer is yes, than we have no problem…. we simply have a solution, you are fighting.
21. October 2010 at 19:09
If the AS curve is vertical (perfect price flexibility), then theory says that a permanent increase in the money supply will cause a permanent increase in P and have zero effect on real or nominal interest rates.
If the AS curve is horizontal (zero price flexibility, even in the long-run) then a permanent increase in the money supply will cause a permanent increase in Y, and the New Keynesian Consumption-Euler equation IS curve tells us there will be zero effect on interest rates too. (Since the real interest rate is positively related to the expected growth rate of consumption, not the level of consumption).
It is only in the case of a temporary increase in Y, due to slow adjustment of prices, that the New Keynesian IS curve tells us that the real interest rate will need to fall to induce people to have temporarily higher consumption.
21. October 2010 at 19:11
Nick Rowe,
De nada. It’s all in the interest of edumification.
By the way I truly enjoy your blog. This is also the first time you’ve ever acknowledged anything I’ve had to say to my knowledge. But that’s alright.
21. October 2010 at 19:13
And even in that latter case the nominal interest rate may still rise when the money supply is increased, because expected inflation is temporarily higher as prices adjust to the new equilibrium.
21. October 2010 at 19:27
Thanks Mark! Ah, but you always have good comments. i couldn’t resist “correcting” your last one, because it so stuck in my mind that the Friedman interlude was such a very brief temporary respite from thinking of monetary policy as interest rates. So you were underselling your point.
Going back to Scott’s “steering wheel” metaphor: imagine you had a car where the power steering was stronger than your hands, and some damn fool mechanic had hooked up the power steering the wrong way around so it turned the steering wheel in the opposite direction to the way you tried to turn it. That’s what monetary policy through interest rate control is like.
21. October 2010 at 19:44
A few reasons:
1) Economists like to think in terms of price and quantity. Interest is the price of money.
2) The focus on the price of money over the quantity of money is because targeting the later rather than the former seems to produce more stability.
3) The price level and output respond at different times to the same shock. This what makes macro, macro. Otherwise it would just be general equilibrium micro. Say’s law would hold even if people hoarded cash because prices would instantly fall in all markets and real cash balances would return to equilibrium.
The idea that there is something funky going on with prices is key to our understanding of the business cycle.
21. October 2010 at 19:52
Karl:
But interest is the price of a loan, not the price of money. The price of anything is the price of money, in terms of that thing.
And targeting the rate of interest, in the sense of holding it constant, produces extreme instability, with either a zero or infinite equilibrium price level. Holding the stock of money constant doesn’t do that.
21. October 2010 at 20:33
Nick Rowe wrote:
“Going back to Scott’s “steering wheel” metaphor: imagine you had a car where the power steering was stronger than your hands, and some damn fool mechanic had hooked up the power steering the wrong way around so it turned the steering wheel in the opposite direction to the way you tried to turn it. That’s what monetary policy through interest rate control is like.”
Actually I feel that the mariner analogies (which I ocassionally read) are more apt. When one steers a sailboat by rudder, one of course turns the tiller the opposite direction one wants to go. If you’ve spend some time doing it, of course it becomes natural.
But naturally when one steers a sailboat there are other tools you should rely on for guiding its course than the rudder. Otherwise you could get yourself into a situation where your rudder is pinned and you’re still off course. Kind of like our economy.
21. October 2010 at 22:32
‘If you create higher NGDP growth expectations, then even if expected inflation doesn’t rise at all, the Wicksellian equilibrium real rate will rise and monetary policy will become more stimulative.’
The way I can think of is supply side effects + money (or velocity of money) money. Money can’t do it by itself, if the monetary authority tries to, it will either lose credibility (and fail) or cause inflation.
Anyway, this also suggests that supply side changes can help NGDP growth without inflation. I didn’t take you for a radical supply sider Scott 😉
22. October 2010 at 00:25
If the problem is that AD is too low because the banks won’t lend the money (or there’s a low velocity in general) due to the fact that people and businesses don’t want to borrow money, because the economy is bad and they are worried about the future; would the problem be solved if only people (and businesses) for some reasson just got more optimistic about the future and therefore borrowed and spent more?
Would that in itself lead to an “easier monetary stance” even if the Fed didn’t do more QE or did antyhing else to help the recovery?
Sorry about that first sentence 🙂
22. October 2010 at 00:27
Economic criticism is an art form and interest rates are its current fashion. When you don’t use interest rates in your economic writings, you are like Schoenberg not using tonality in his musical compositions.
22. October 2010 at 00:53
Karl
1) Economists like to think in terms of price and quantity. Interest is the price of money.
I literally yelled out when I read that. That this is the automatic notion of what interest rates are seems to be behind so much that is wrong with the thinking of so many economists. Starting with “low interests are a sign of loose money”. Because, if the price is low, the natural assumption is that supply is high: hence low interest rates = loose (in plentiful supply) money.
It is, of course, just not so.
So, what Nick said. Interest rates are an indicator of future expectations about the price of money since that is one of the constituent elements of interest rates (along with small matters such as risk): that is as close as interest rates get to being the price of money.
22. October 2010 at 02:53
Anyway, food inflation is now rising very quickly. 🙁
22. October 2010 at 03:26
I quite agree that adjusting interest rates is largely waste of time. Coincidentally I argued that point on my own blog about a week ago. Specifically I argued that: 1, if short term rates had a quick and significant effect on long term rates, that would be a point in favour of adjusting interest rates, but that short/long effect is feeble, 2, in that low interest rates encourage investment in a recession this is exactly what is not needed because in a recession there is a load of capital equipment lying idle, 3, altering interest rates is distortionary in that the ploy works only via entities that are significantly reliant on variable rate loans, 4, credit card spending is not much influenced by short term changes in interest rates.
22. October 2010 at 03:35
Scott, you see macroeconomics from a Wizard of Oz perspective – what the guy behind the curtain needs to do with the dials on his console to make everybody happy. Policy should be the last topic in a macroeconomics textbook and not the first. Interest rates are one of the most concrete and relevant macroeconomic phenomena. Perhaps, interest rates belong in chapter one. When we get to the last chapter we can look at what policy can and should do to or with interest rates either as an unintended result or as policy variable and what the consequences are. Along the way we look at what money is and how it is created. We look at phenomena like the general price level, prosperity, unemployment and the process by which unexpected shocks reverberate through the economy. These are abstractions that are difficult to measure but none the less correspond to experienced reality. We look at the general conditions that promote prosperity – secure property rights, the rule of law, freedom to contract, cultural factors. We look at the human life cycle and the biologically driven patterns of consumption and work. We look at government as a part of the economy with certain peculiar characteristics – government as a part of the system rather than as the Wizard of Oz. We look at all these things from an international perspective.
When we come to policy, macroeconomics becomes speculative. There are a few things we know – probably enough to prevent a policy-generated catastrophe. Opinions differ and approaches differ when we get to the details. Here is where you get to privilege your own approach and opinions.
22. October 2010 at 03:49
“But interest is the price of a loan, not the price of money. The price of anything is the price of money, in terms of that thing.”
That’s not right, Nick.
I can price a car in terms of money, or I can price it in terms of a loan, which also has a money price. The money price of a loan is its present value taking into account prospective interest earned. So I can price a car in terms of a loan by pricing it in terms of the money price of a loan.
Similarly, I can price money in terms of cars or loans.
22. October 2010 at 04:08
To defend his point that austerity now is the wrong tactic, Krugman relies on interest rates:
“Over-reliance on the financial industry largely explains why Britain, which came into the crisis with relatively low public debt, has seen its budget deficit soar to 11 percent of G.D.P. “” slightly worse than the U.S. deficit. And there’s no question that Britain will eventually need to balance its books with spending cuts and tax increases.
The operative word here should, however, be “eventually.” Fiscal austerity will depress the economy further unless it can be offset by a fall in interest rates. Right now, interest rates in Britain, as in America, are already very low, with little room to fall further. The sensible thing, then, is to devise a plan for putting the nation’s fiscal house in order, while waiting until a solid economic recovery is under way before wielding the ax”.
22. October 2010 at 05:36
Scott:
Interest rates play an important role in the economy.
They coordinate of the production of goods with consumption through time.
Real interest rates are important because that is what creates the incentives to provide for the coordination. It is the relevant _relative_ price.
Good macroeconomic theory should roughly understand the macroeconomy when there is no monetary disturbance. And then you look at how monetary disturbances mess it up.
As you well know, your paradigm of monetary theory falls apart with pure inside private monies.
Thinking about monetary policy as if _all_ money is zero-interest hand to hand currency that is dropped from helicopters–which is your basic approach–captures some aspects of the status quo, and some important aspects of the medium of exchange. But it doesn’t capture all of them.
The pure private credit money system captures other aspects of the status quo. I know for a fact that you have trouble grasping it.
The Wickellian approach of the natural (or neutral) interest rate is very helpful.
I entirely agree that the focus on interest rate targeting is a mistake. I think most economists do it because the central bankers really, really, want to target interest rates.
As I have explained before, I think your approach to macro, explained above, maybe help explain how monetary policy can stabliize hours worked (or something,) but it provides next to no insight on the macro aspects of the market order.
22. October 2010 at 07:07
Nick said:
“Suppose central banks bought bricks in open market operations. Would we have a story of the transmission mechanism in which the price of bricks played the leading role?”
We would have a story of the transmission mechanism in which change in the monetary base produces changes in the short term interest rates. When Bank of China increases monetary base by purchasing US Treasuries, I guess their newspapers discuss it in terms of changed Chinese short term interest rates; Chinese media mostly ignores changes in the price of bricks/US Treasuries.
22. October 2010 at 07:08
“If I wrote a macro textbook, I would try to avoid any mention of interest rates or inflation. The Fisher equation would use expected NGDP growth. The AS/AD model would use hours worked as the real variable and NGDP as the nominal variable.”
I kind of like that. I’m nowhere near ready to give up interest rates (and how could you have a Fisher equation without them?), but I have long thought that measuring inflation might be more trouble than it’s worth. I would like to give up trying to distinguish between inflation and productivity growth, because the distinction is often fairly arbitrary anyhow.
“But if the policy is expected to work, wouldn’t it boost long term rates?”
In theory (and I personally think it would be a good idea) the Fed could follow a policy of pegging long-term rates (just as it has followed a policy of pegging short-term rates). In that case, the policy could obviously work without boosting long-term rates, although it might destroy the liquidity of the long-term bond market (and obviously the Fed would have to be concerned about inflation if it were too slow to raise the peg once it started to work). In principle, the same result could be obtained with a quantitative target, although the target might have to be much larger than what is now contemplated.
“Some even argued that the Fed would have to boost inflation expectations to 6% in order to get a robust recovery, forgetting that this view directly conflicts with another key assumption of Keynesian macroeconomics””that the SRAS is very flat when unemployment is high.”
That is not a contradiction; it simply means that some positions of the AD curve (and therefore some points on the SRAS curve) are infeasible. It’s what Paul Krugman once called (in reference to Japan, I think) “the excluded middle.” You could think of the SRAS curve as a flat section and a vertical section with a very large gap inbetween.
Some of my commenters argued you couldn’t raise NGDP without first creating inflation expectations, which would lower the real rate of interest. But that’s not necessary at all. If you create higher NGDP growth expectations, then even if expected inflation doesn’t rise at all, the Wicksellian equilibrium real rate will rise and monetary policy will become more stimulative.
It depends. The Wicksellian equilibrium rate will rise, but it may not rise enough, so it may be necessary to have more inflation anyhow.
“When we explain why a big crop of apples makes NGDP in apple terms rise sharply, we don’t resort to convoluted explanations involving an interest rate transmission mechanism. Why then, when there’s a big increase in the supply of money, do we think it will only affect nominal GDP in dollar terms via some sort of interest rate transmission mechanism?”
A big crop of apples is analogous to a helicopter drop of money. Even a helicopter drop might not be very effective on the margin (i.e., if it was not very large) because the demand for money may be highly elastic in the current range. (People have arguments why the demand for money might be so elastic, and there are no analogous arguments for apples.) In any case, a helicopter drop is not the policy under consideration: the analogy to QE would be a big crop of apples offset by a smaller than usual crop of oranges; if apples and oranges are close substitutes, this wouldn’t have a large effect on the apple-denominated GDP. Interest rates are just another way of talking about the degree substitutability between apples and oranges (i.e., money and bonds).
22. October 2010 at 07:28
Nick said:
“And targeting the rate of interest, in the sense of holding it constant, produces extreme instability, with either a zero or infinite equilibrium price level. Holding the stock of money constant doesn’t do that.”
Broad stock of money is unobservable. If you want stabilize the broad stock of money, you have to target interest rates. In September 2008 Bernanke targeted the quantity of monetary base. It was a mistake, AD fell sharply. On October 13, 2008 Bernanke switched to interest rate targeting and said the quantity of monetary base will be solely determined by the LOLR interest rate. It was the first green shoot.
22. October 2010 at 07:32
Scott & Bill – Interest rates matter when it comes to how the markets work every day. The EMH debate proves the point. Hedge funds, which controls a significant portion of the money supply uses interest rates every day to drive trading. I would argue we two economies that while linked for inflows/outflows purposes but remain largely uncoupled and these two have created two different intermediate currencies. The “finance economy” comprising hedge funds etc. uses an “interest currency” while the “production economy” made up of operating companies relies upon a “productivity currency”. The finance economy uses EMH latency affects largely driven by interest rates to make money in “episodic transactional” environments. That capital largely stays in place growing or shrinking, but largely we see very little redemption for the purpose of being used in the “Production Economy”. The Production Economy which uses “Productivity Currency” is just the opposite. It relies the systemic increases in intrinsic value realized from output vs. input cost created on even small improvements in productivity that produce scalable and repeatable economic returns. Here interest rates don’t matter as much because low capital (interest) cost vs. the value created on even small improvements in productivity. So in a sense Interest Rates and Productivity Rates are interchangeable “measurements” for a particular economic environment. Today, investors put their money to work where it will get the highest return. So if you are the treasurer of an operating company and you can improve productivity by investing higher returns than that cash sitting with a hedge fund…. You get my point.
22. October 2010 at 08:50
When Bank of China increases monetary base by purchasing US Treasuries, I guess their newspapers discuss it in terms of changed Chinese short term interest rates;
BOC purchases of Tpaper have no direct effect on monetary base. Only FED desk operations do. BOC purchases from Treasury through FED, are sequestered, and show up in TIC flows. What effect do such purchases have? Surprisingly indeterminate because it depends on other factors like state of the US budget and psychology of Tpaper market primarily wagged by inflationary expectations.
When Temporary is closed, as it has been for 2 years, FED, desk, can only effect changes in reserves. They have no other power. For example, they have no control over the quantity of currency in circulation. This quantity is exclusively determined by public preference to hold money balances where such currency quantities are charged against member bank reserve position. When Temporary is open during a demand regime for loanable funds FED then has limited power to influence economic activity through rate fixing. When Temporary is closed during a supply regime with loans running off, reserve creation is like pushing on a string.
22. October 2010 at 09:50
The reason is obviously historical: Central banks used to be banks (except the Fed, which is formally a consortium of banks). Banks lend money for some interest rate. Central banks -unfortunately- kept the habit to act as if they were still banks by advertising the interest rate.
Macroeconomists and press have just the same bias.
22. October 2010 at 10:25
@ahhaha
I was talking about Chinese monetary base.
22. October 2010 at 10:54
http://www.washingtonpost.com/wp-dyn/content/article/2010/10/22/AR2010102202509.html?hpid=topnews
22. October 2010 at 11:29
BOC purchases of US Tpaper have no effect on Chinese monetary base. BOC doesn’t even need to sterilize. It’s merely a matter of refactoring,. that is, taking trade dollars and using them to buy Tpaper.
22. October 2010 at 11:38
Blinder says, “Now they’re planning to buy long-term Treasury bonds that are already in high demand. They’re creating liquidity, in other words, where it already exists.”
Not correct. They aren’t creating “liquidity”. This is why I pointed that FED can only create reserves. In contrast, any sense of “liquidity” must involve a sense of “money”, i.e., that which is fungible since fuungibility is the notion Blinder implies.
FED’s current operations which are referred to somewhat incorrectly as “POMO” aren’t actually “O”. FED buys Permanent, Tnotes, not across the board, but selectively. An examination of the Tnote CUSIPs shows that these notes are the ones associated with the conveyor belt rescue facilities and their exchange for toxics. By default, this is quasi sterilization in that the only effect is to create yet more reserves, reserves which banks can’t and won’t use to support C&I. Thus monmetary policy is DOA. Need appropriate fiscal policy to get out this trap.
22. October 2010 at 11:39
Blinder says, “Now they’re planning to buy long-term Treasury bonds that are already in high demand. They’re creating liquidity, in other words, where it already exists.”
Not correct. They aren’t creating “liquidity”. This is why I pointed that FED can only create reserves. In contrast, any sense of “liquidity” must involve a sense of “money”, i.e., that which is fungible since fungibility is the notion Blinder implies.
FED’s current operations which are referred to somewhat incorrectly as “POMO” aren’t actually “O”. FED buys Permanent, Tnotes, not across the board, but selectively. An examination of the Tnote CUSIPs shows that these notes are the ones associated with the conveyor belt rescue facilities and their exchange for toxics. By default, this is quasi sterilization in that the only effect is to create yet more reserves, reserves which banks can’t and won’t use to support C&I. Thus monmetary policy is DOA. Need appropriate fiscal policy to get out this trap.
22. October 2010 at 12:09
Today in class I showed Episode 3 of Friedman’s Free to Choose explaining the causes and consequences of the Great Depression (it’s free via Google Video). It was good to illustrate two points:
1. Money does matter.
2. If the U.K. pursues fiscal contraction its bank can still prevent a recession even if it is facing the zero lower bound.
3. We haven’t learned anything since 1930, the same battles are being fought.
22. October 2010 at 12:19
Randall, that was pretty incoherent, and filled with fundamental confusions. For example, it’s not even meaningful to talk about an “interest rate currency”, an interest rate is a property of a good, not a good itself. One cannot hold 6 interest rates. Also, hedge funds do not “control a significant portion of the money supply”.
22. October 2010 at 12:24
Here’s a finer historical simplification and answer to the headline question:
Age of Strong
Age of McChesney-Martin
Age of Volcker-Greenspan-Bernanke
Probably none of you know about the first two. In spite of criticism arising from presumably mishandling monetyary policy in 1930 by none other than Milty, Strong was capable and disciplined. Embracing greater disc9ipline McChesney-Martin believed that it was FED’s role to allow a free market in money determine its price, although he never stated this overtly, but acted as though that was critical. The result was that the monetary base hardly grew in the ’50s, but the economy did very well. Enter the school of Keynesiian demand management eternal prosperity insurers who hate discipline. “A little inflation like a little drink never hurt anyone” devolved into “We need to engineer inflation to survive”. Sounds like the economics of ruination under the banner of pretense to knowledge.
22. October 2010 at 14:34
Why is Scott Sumner so obsessed with NGDP?
I can understand interest rate mechanisms… I’m not sure I understand Scott’s NGDP mechanism.
For a firm, the interest rate represents an opportunity cost of investment. If an investment will only pay a return less than the interest rate, then if you want to maximize profit you shouldn’t make that investment. For a household, the interest rate is the opportunity cost of consumption.
So no matter who you are, the interest rate should be a relevant variable in your decision making. How is NGDP a relevant figure?
22. October 2010 at 15:59
This is where I came in…to internet discussions of economics. About 15 years ago an academic economist was asking for contributions to ‘myths of economics’. Mine was, ‘interest rates are the price of money’.
Nick Rowe and Lorenzo are correct, interest rates are the price of renting money; no ownership, temporary use only. The price of (buying) money is whatever you give up in exchange for it. Your labor, say. Or, if you’re a car dealer, the car you exchange for it (whether the new title holder borrows the money or not).
The important distinction being that the two prices often move in opposite directions. When the price of buying money is declining–as it was during the inflationary 70s–the price of borrowing, or renting, money increases. What Scott calls, I believe, the Fisher Effect (and I learned as Gibson’s Paradox).
22. October 2010 at 16:04
Matt, You sure know a lot about apples.
Nick, You said;
“Scott: I think this is the right question to ask. But I think your discussion of the slope of the SRAS curve is a red herring. The question is: why should the transmission mechanism from money to the quantity of output demanded, at a given price level, (i.e. the rightward shift in the AD curve at a given P) be seen in terms of interest rates? That question is independent of whether the SRAS curve is horizontal or vertical.”
You might be right, but I don’t think it is a complete red herring. It shows two aspects of the Keynesian model are incompatible. One aspect is the view that really low real rates are required for a robust recovery, and the other is the widely held view that the SRAS curve is fairly flat at high unemployment. We definitely need more AD to get a robust recovery, but whether our need for much more AD implies a need for much more inflation, depends on the slope of the SRAS. Certainly the slope of the SRAS should have nothing to do with whether AD shifts, but it has a lot to do with whether a given shift in AD implies high inflation.
I completely agree about the misleading term “central bank.”
If I ever get caught counterfeiting, I’ll tell the government that I am a central bank.
Mark, Thanks for the support. I suppose more of my blog seeps into my course than I realize. Indeed I suppose I also steal an occasional idea from people like Nick.
Nick, I think of Wicksell as the first major new Keynesian (using an interest rate-oriented approach.) But like everything in economic history, I assume there is always someone earlier.
Morgan, Isn’t it the foreclosure mess, not the Fed, holding things up?
Nick, you said:
“It is only in the case of a temporary increase in Y, due to slow adjustment of prices, that the New Keynesian IS curve tells us that the real interest rate will need to fall to induce people to have temporarily higher consumption.”
This isn’t my area of expertise, but Robert King claimed (in 1993) that monetary stimulus could raise both nominal and real rates in a rational expectations new Keynesian model. I presume the rise of real rates resulted from much higher expectations for real growth (let’s say due to wage and price stickiness.) Have you run across this argument?
Karl Smith, You said;
“1) Economists like to think in terms of price and quantity. Interest is the price of money.”
I think you are right about economists, but I’ve never liked that definition of the price of money. I’d prefer to call 1/P the price of money, and call the interest rate the rental cost of money. Oops, so much for my goal of macro without the price level.
You said;
“2) The focus on the price of money over the quantity of money is because targeting the later rather than the former seems to produce more stability.”
I agree that interest rate targeting often works pretty well. What puzzles me is why people like John Cochrane are only now getting into CPI futures targeting. This idea (which I’ve been pushing for 24 years) fits naturally with ratex and the EMH, which are core concepts in right wing macro.
I agree with your third point about wage and price stickiness being the key to macro, but that’s also true in the simplified macro that I’d like to teach. Macro w/o interest rates. The “liquidity effect” would be replaced with the “asset price effect”. In the short run (due to sticky prices) monetary shocks would raise the price of assets like stocks and commodities and (over a slightly longer period) commercial real estate.
Doc Merlin, I’m a moderate supply-sider. I do think supply shocks can indirectly boost AD, by raising the real return on capital.
more to come . . .
22. October 2010 at 16:08
‘So no matter who you are, the interest rate should be a relevant variable in your decision making. How is NGDP a relevant figure?’
Andrew, Scott is a MACRO economist, so he studies things like GROSS domestic product. The opportunity cost of household consumption is micro.
22. October 2010 at 16:29
Nick and Lorenzo:
So I should say interest is the opportunity cost of holding money. What you give up. I suppose that’s not technically a price in the sense of a rate of exchange.
22. October 2010 at 16:59
Patrick, I think you misunderstand my comment. Obviously figuring out what determines NGDP is of interest to a macroeconomist.
I am merely confused about the mechanisms in Scott’s model, and was hoping someone might be able to explain it to me. My macro background is pretty limited. When it comes to business cycle models, all we got to was IS-LM and RBC models (as I said, pretty limited!).
Let me lay out more explicitly what I don’t understand:
1. “The Fisher equation would use expected NGDP growth.”
I’m not sure what this means. The Fisher equation is r = i – pe, where r is the real interest rate, i is the nominal rate, and pe is the expected inflation rate. What is the NGDP equivalent?
2. “The AS/AD model would use hours worked as the real variable and NGDP as the nominal variable.”
So assuming AS is upward sloping and AD is downward sloping in this space, how are those curves derived? What are their underlying assumptions?
3. “More cash would raise expected future NGDP.”
Why? I know that MV = PY, but that doesn’t mean agents couldn’t expect V to fall in response to increases in M.
4. “This would raise the current price of stocks, commodities and real estate.”
Again, why?
5. “The higher asset prices would tend to raise current AD. More nominal spending, when combined with sticky wages and prices, would boost output.”
This part I understand.
By the way, I didn’t mean to imply any disrespect for Scott or his views, I’m just someone trying to understand.
22. October 2010 at 17:47
Mattias, Yes, and that’s why an explicit (and higher) P or NGDP target is useful–it should make people more optimistic for any given money supply.
woupiestek, The Schoenberg of economics? I’ll take that as a compliment (even though I am ignorant of modern music.)
Ralph, In my view the markets paid attention to when the Fed moved interest rates because they saw it as an indication of future trends in the money supply.
Charles, Those are all good points. I’m actually thinking of a subset of macro–business cycle theory.
JKH, When I ask the price of something, I generally want the cash price. I get annoyed when used car dealers say “it’s only $99/month” I want the cash price.
Marcus, I wonder why Krugman doesn’t call on Britain to devalue the pound?
Bill, I know of monetary systems based on fiat money, and those based on commodities like gold and silver. Are there any other monetary systems that I should be aware of?
I suppose you are right about interest rates and inter-temporal choice. I should have said business cycle theory. The AS/AD model obviously relates to business cycle theory.
Andy, Yes, I wasn’t thinking when I discussed the Fisher equation w/o interest rates. What I meant was a book w/o inflation, and a business cycle model w/o interest rates.
I agree with you about inflation, it’s an arbitrary concept. You said:
“That is not a contradiction; it simply means that some positions of the AD curve (and therefore some points on the SRAS curve) are infeasible. It’s what Paul Krugman once called (in reference to Japan, I think) “the excluded middle.” You could think of the SRAS curve as a flat section and a vertical section with a very large gap in between.”
I disagree. There is no infeasible AD position. Use the monetarist definition of AD as a given NGDP (a hyperbola). Then do NGDP futures targeting. The market sets the monetary base at the level expected to produce on-target NGDP growth, for any central bank target. There is no excluded middle. Krugman thinks there is one, because he’s completely relying on the interest rate approach to AD.
You said;
“It depends. The Wicksellian equilibrium rate will rise, but it may not rise enough, so it may be necessary to have more inflation anyhow.”
Yes, but that’s exactly where the SRAS comes in. (I’m still responding to Nick here.) Whether you are able to get enough growth without inflation depends on the slope of the SRAS curve. If it’s very flat, perhaps you can get 8% real GDP growth with 1% inflation. I’d say that’s enough RGDP growth to get you out of a liquidity trap, and there’d be little inflation. If the SRAS was steeper, then I agree the required NGDP growth rate might lead to somewhat higher inflation. My assertion was based on the assumption that SRAS is now fairly flat.
You misunderstood my apple analogy. I wasn’t trying to argue against liquidity traps, but rather that we don’t normally use interest rates to explain why a big crop of apples raises NGDP in apple terms. I was criticizing the fact that even when we are not in a liquidity trap, economists insist on explaining the monetary transmission mechanism using interest rates. Even when cash and T-bills are not close substitutes.
Randall, Yes, interest rates matter in finance. I should have said “business cycle theory,” not “macro.”
ahhaha, Are you talking about US or Chinese monetary policy? You were responding to someone talking about China.
Jean, I agree.
123, Yes, I thought you meant China.
JimP, Thanks–I don’t think fiscal stimulus will occur.
ahhaha, We need appropriate monetary policy, not fiscal policy.
JTapp, Yes, I see the same mistakes as well.
ahhaha, You said;
“Probably none of you know about the first two. In spite of criticism arising from presumably mishandling monetyary policy in 1930 by none other than Milty, Strong was capable and disciplined.”
I’m not sure what’s funnier; that you think none of us have heard of Strong, or that you think he was in charge of the Fed in 1930–when he was dead.
NGDP, Do you think business do the same amount of investment if they think NGDP will grow at 5%, as they would if they thought NGDP would fall at 5%.
Patrick, I agree about the price of money. Although interestingly the Gibson Paradox turns out to be slightly different from the Fisher effect. With the Gibson paradox, high prices are associated with high interest rates. With the Fisher effect high inflation rates are associated with high interest rates.
22. October 2010 at 18:02
Karl Smith, An exchange rate analogy might be useful. If the British pound price of dollars in .50, then the dollar price of pounds is 2.00.
If the dollar price of a pound of butter is $2, then the butter price of a dollar is 1/2.
So for the overall value of money with respect to all goods, you could say it’s 1/P, where P is the price level.
Andrew, You are right that my Fisher comment was weird, as it includes an interest rate. I meant that I’d replace the inflation term with the expected NGDP growth term. I’ll do an update.
For AS/AD, I’d have a horizontal AD curve representing a given NGDP, and an upward-sloping SRAS curve representing the response of hours worked to changes in NGDP.
3. Money is neutral in the long run, so an increase in M that is expected to be permanent should not impact V in the long run.
4. I believe that stocks, commodities and RE prices fall when NGDP is expected to fall. That’s mostly based on my study of economic history–it’s a theory driven by empirical evidence. But the theory would be that a fall in NGDP reduces real corporate profits, and also reduces new investment in buildings (because existing buildings are not fully utilized as economic output falls.) That reduces the value of current buildings. Commodities are generally recognized as a very cyclical asset, so I don’t think that is very controversial.
If wages and prices are completely flexible, then asset prices also respond in the same way, although in that case there is no change in real asset prices, only nominal prices.
22. October 2010 at 18:40
I agree that interest rate targeting often works pretty well.
This view which is universally held causes all the problems. The staff at FED when pushed about the need for free market price discovery don’t think a free market in money is appropriate. For some reason they think the 1913 charter calls for control related to “lender of last resort” when in fact, it doesn’t. I ask you, how can the equilibrium rate ever be discovered when FED fixes the price? Since it can’t and since that means such fixing causes the market for money always to be in disequilibrium,. what are the consequences for this deceptive illusion of control?
The asnwer is simple. Increased amplitude of both economic and financial excesses including hidden ones like the one that led to the 2008 financial crisis. FED allowed banks to exceed reserve requirements because their operations fixing the rate through Temporary obscured how the created funds, the RP free float, were being used. Some of those funds were used indirectly to support excessive leverage in the RE market. That wouldn’t have been the case with a free market in money which reacts to every infinitesimal factor however far removed since the marginal lenders are often the perpetrators of excess. In effect, FED, trying to manage economic outcomes lost sight of the degree to which Temporary was redirected into the RE, broke the implicit connection between borrowers and lenders which sent the true spread much farther apart than was reflected in the usual measures.
Are there times when rate targeting doesn’t embed distortions? There never are. FED always diverges from equilibrium until they lose control, and control then defaults to the implicit free market which has to rise up to protect the very money FED creates and uses. Specifically, lenders revolt and FED must get out of their way This is what happened in 1979 but you won’t read that story in any depiction of financial historical.
22. October 2010 at 21:32
The ‘interest-rate’ transmission mechanism appears in the 1912 edition of Mises’ Money and Credit which is presented as accepted fact. Indeed, it assuredly appears in Wicksell’s Interest and Prices from 1898. So, I’d venture that it well predates Keynes!
I think you need to step back and consider the history of monetary policy theory. Going into the 1800s we had the Bullionist debates, and the idea that Adam Smith’s real bills doctrine extended to central bank. i.e., that the issuance of currency should be fully elastic to the demand for credit by real bills.
One of the monographs arguing against the real bills doctrine was that of Thornton’s Paper Credit. Certainly that must be the genesis, because the essential claim there is that discounting by the central bank influences the terms of credit and so increases or decreases the equilibrium amount of money. i.e., accommodate breeds new demand which in turn causes further accommodation and so forth. What matters is how the market-rate compares to the expected rate of profit.
So the two-rate model, was the first and strongest prong of the rebuttal against the real bills doctrine. It was understood then, that if they CB wanted to pursue a policy of an elastic currency, it would need to stabilize the supply of credit with the interest-rate at the natural-rate.
Thorton was later involved a parliamentary committee on the Bank of England–The Bullion committee, investigating the ‘high price of bullion’. “The Bullion Report”, they issued, laid out fairly cleanly what is now referred to as the first-round credit channel and the second-round money channel. This is only in 1810.
These ideas are very old. They stem from a era was money was gold or silver, and “notes” literally were credit. The trouble then identified that notes circulated as money substitutes and affected the price-level. Ergo, since notes are now money, we still talk about interest-rates.
22. October 2010 at 23:13
Thanks for the clarification.
It just seems that the mood in the US seems to be so much more pessimistic than in Sweden and many other countries. I have a theory that sometimes there’s a limit to how long people can be pessimistic without getting tired of it and instead sort of continuing with their lives (which usually means buying more stuff again). I think we have already reached that point in Sweden.
23. October 2010 at 03:11
Scott, where are the markets that predict the future monetary base, MZM, M2? Economic agents don’t really need such information, they are obsessed with markets that predict future real and nominal interest rates. No wonder many economists share this obsession.
23. October 2010 at 06:02
ahhaha, You misunderstood me. I said it “often” works well. It sometimes works horribly, which is why I oppose interest rate targeting.
Jon, Thanks for all that info. It was also my impression that the interest rate channel goes far back in history.
Mattias, Sweden has a much more competent government than we do–that explains part of the difference. I assume people in Sweden were depressed in the early 1990s.
123, I think what people really care about is future NGDP—which is roughly what businessmen mean when they talk about “confidence.”
As you know, I pay no attention to the aggregates.
BTW, if you told people in 2007 that in 2009 and 2010 the Fed would be holding rates at 0.25% (and real rates would be negative), would that foreknowledge of “easy money” have made people optimistic? Bullish about new investments?
23. October 2010 at 08:52
Scott:
You should know about likely future monetary systems that have no zero-interest hand-to-hand currency, but in which all money is interest bearing deposits.
Further, you should be especially interested in monetary systems in which nominal GDP is targeted, so that the quantity of interest bearing deposits must be adjusted to meet the demand to hold them given that target. And, of course, when the demand to hold them depends on the interest rate paid on them. (This would be as opposed to a scheme where the quantity of those deposits could be increased based upon desired government spending or else according to a fixed growth rule with the interest rate set arbitarily, perhaps at zero.)
Under those very realistic institutions, the quantity of money would represent borrowing by the issuer, which might be the government. Of course, because most deposits are privately-issued today, there is no reason to assume that all, most, or any would be issued by the government. Further, they would all bear interest.
Go ahead and imagine that the government issued interest bearing deposits are “base” money and all of the private interest bearing deposits are redeemable in it. Plausible. Now, suppose the demand for base money falls-alot, say 90%, because people want to hold the privately issued deposits.
With a “fiat” mentality, this is inflationary. But, we have nominal income targeting. The government must decrease the quantity of its base money by 90%. How does it do that? A sudden huge budget surplus? Come on! It sells nonmonetary, interest bearing bonds so that the quantity of its money drops enough.. etc. Of course, it could make its money more competitive by paying higher interest.
But, notice– interest rates and credit are clearly integrated in the process. What would those who buy the nonmonetary government bonds have done with the money? Lend it to the private sector?
Treating the money issuer like a bank (it issues money and perhaps buys nonmonetary, interest bearing government bonds) is perfectly sensible. Having the fiscal arm of the government always borrow by issuing bonds, and the money issuing arm buying some of those bonds, collecting interest from the fiscal dept, and passing that on to those who hold the money issued by that arm makes perfect sense.
While that isn’t the system we have today, the system we have to day is closer that likely future system than a gold coin standard or a system where money is all zero-interest hand-to-hand currency printed up and then spent by the government, with its value determined (the price level determined) by the quantity issued and the real demand to hold money.
Money and credit are tied together. The Fed is commmitted to an inflation target, and doesn’t let inflation or the price level get determed by a given quantity of money. The quantity of money adjusts to reach the inflation target. Currency is an afterthought, with nearly all payments made by electronic or check transfer of some sort of deposit account. Most of those are private. All of those deposit accounts are matched by some kind of financial instrument held on the balance sheet of some kind of financial intermediary.
This means they are borrowing–they have to be ready to pull it out of circulation–pay it back. And they may have to pay nearly all of it back. There is no reason to expect that the demand for the monetary base will always grow or that there will be more than an arbitrarily small amount demanded in 20 years.
The Fed does operate like a bank, with monetary liabilities and holding assets. You can call that an illusion if you wish, and treat the issue of base money as if it is printed up, but not if there is anything other than a quantity rule or an irresponsible “spend what we want” rule. If we have a NGDP rule, or what we actually have, then they have to be ready to “pay it back,” that is, pull it out of circulation based upon demand.
All of this has practical, analytic consequences too. Your fancy schemes to reduce excess reserves so that currency will somehow flow into the economy and then currency expenditures will raise NGDP just don’t add up.
Of course, I agree that interest on reserves should be negative now, that interest rate targeting is a mistake, and that a committment to get NGDP back to a reasonable growth path should raise market interest rates.
But a macroeconomics that ignores interest rates would be very much in error. But then, I think “supply side” economics should be an important part of macroeconomics as well. In my view, a good macro text would be mostly supply side, including saving, investment and interest rates. The problems created by monetary disequilibrium and how alternative monetary institutions can worsen or minimize them should be just a part of macroeconomics. I think the status quo, likely future monetary institutions, and desirable ones, all closely integrate money and credit markets. And so, a macro text that singlemindedly focused on how ngdp targeting would avoid undesirable fluctuations in employment would be awful. And one that treats money as if it were paper gold would be bad as well.
23. October 2010 at 10:50
How about this: a macroeconomics free of intertemporal coordination across production processes, credit, savings, consumption, and investment … oh, that’s right, we already have that.
It’s called mainstream academic economics …
23. October 2010 at 11:35
ahhaha,
About that Age of Strong, see: http://www.themoneyillusion.com/?p=6061
23. October 2010 at 13:04
We don’t have to imagine big changes only ONE;
Most Merchants accept more than one currency.
Immediately, you choose your poison and demand for a given currency is based on who does the best job of convincing you they will not PRINT.
23. October 2010 at 13:45
In today’s WSJ Stiglitz attacks quantitative easing.
“Why Easier Money Won’t Work”
Because interest rates can’t go lower, competitive devaluation is beggar-thy-neighbor policy, QE won’t help lending by small banks to small businesses to create jobs, and a bond bubble is bad.
Well, he and Krugman have separated themselves on this issue. On video Paul says we need $8 trillion to $10 trllion of QE (and “China is really the bad guy in all this”).
23. October 2010 at 18:22
I think Stiglitz reads Simon Ward and realizes (like Matty and Ezra are figuring out) hitching your liberal strategy to QE is bad because 2% CPI is right around the corner.
http://www.moneymovesmarkets.com/journal/2010/10/22/its-the-velocity-stupid.html
23. October 2010 at 19:03
I’ve been reflecting on the “if I wrote a textbook” issue. At the University of Delaware there is an unnecessary plethora of Principles of Microeconomics books. Two of our professors (unnamed on purpose) use their own unpublished textbooks. To this add Bade and Parkin, Krugman, Mankiw etc. etc. As a result the group tutorial in the principles of microeconomics held at UD (hosted by me) is a cacophony. I usually try to go round the tutorial room spinning each subgroup in discussion as I make the rounds.
Oh it makes me mad!!!!(approximately 3:30 in)
http://www.youtube.com/v/Z9VLwV48OHs&autoplay=1&fs=1&autoplay=1
The same exact material presented in nauseatingly different detail! Confusing to students in the extreme!
Where do we actually need more textbooks? Macro! Most macro texts are dreadful. Tabarrok and Cowen is good at the principles level of course, but other than that I am hard pressed to think of any level of macro text (other than maybe Mishkin and Blanchard) that I can claim to be good.
23. October 2010 at 19:11
While interest rates are obviously important for intertemporal coordination, it seems that much can be accurately explained without explicit mention of them. But contemporary economics uses interest rates as the locus of all understanding and explanation in macroeconomics, when sometimes a different approach (or locus, such as NGDP) would improve and clarify understanding.
That is what I took Scott’s point to be, but perhaps I just misunderstood.
23. October 2010 at 23:40
Scott,
You misunderstood me. Fixing interest rates, like fixing prices in markets, never works.
The usual counter argument to free market in money is based on a presumed need for concession to control. They and FED believe that FED can’t be lender of last resort if the free market is given sway. A free market in money need not operate to the exclusion of any and all authoritative intervenion. What we have had since the late ’60s is FED price fixing, every day through Temporary, and this continued and exclusive interventionism embeds all financial and economic dislocatioins.
As for my comments about BOC, People’s Bank of China, when it converts its trade dollars into Tpaper, those dollars never leave the US. There’s no effect on the Chinese monetary base.
As for my comment about the Era of Strong and how it foundered in 1930, didn’t imply that Strong still had the chair or was alive. His style of management was still in place, and that was part of the problem causing quixotic changes in policy in 1930.
I’m done here.
Good bye.
24. October 2010 at 05:27
ahhaha,
I especially believe that interest rate smoothing is a counter-productive aspect of interest rate targeting.
At times, I wonder if the Fed’s actual goal is minimizing fluctuations in short term interest rates for the benefit of
security traders on Wall Street.
That this might lead to macroeconomic disaster is a constraint–darn it, the just have to change those short term interest rates or else depression or hyperinflation result. So, how do we minimize the fluctuations in short term interest rates subject to the constraint that these efforts don’t create more that acceptable levels of macroeconomic disruption.
This speculation is at least partly driven by the fact that the Fed has operated for years (leaving aside the last three) by making short term overnight loans to security traders secured by an unspecified variety of government bonds.
24. October 2010 at 05:48
Bill, take look at the intraday volatility of effective Fed funds rate in September-October 2008, and you will see that “smooth” interest rates were not the problem in September-October 2008.
24. October 2010 at 05:53
Scott, you said:
” I think what people really care about is future NGDP””which is roughly what businessmen mean when they talk about “confidence.”
As you know, I pay no attention to the aggregates.”
People care about future NGDP and RGDP. Beyond that, they care about various interest rates, and those interest rates are very important inputs in the forecasting process, no wonder macroeconomists are obsessed with them.
24. October 2010 at 06:04
You misunderstood me. Fixing interest rates, like fixing prices in markets, never works.
… like any market, fixing prices affects supply and demand of the relevant good. Which is exactly the intention here, is it not?
24. October 2010 at 06:23
@ahaha:
You have to fix one price, otherwise the money has no value. Two prices is of course too much.
That can be gold, an other currency, CPI or NGDP. Fixing interest rates is one (dirty) way to do that. Fixing a money supply would work if velocity of circulation were constant.
24. October 2010 at 06:57
Bill, I am a pragmatist, and prefer to model monetary systems that actually exist, or have existed, in the real world. None of those involve interest-bearing currency.
It is possible that at some point currency will pay interest, and that the Fed will use adjustments in the interest rate on currency to determine the price level. But even in that case the market interest rate for private sector assets may play little or no role in business cycle theory. For instance, it is very possible that when the Fed sets its policy rate higher that expected, market interest rates fall. Indeed this is precisely what happened in December 2007. This tells me that market interest rates are a symptom of business cycle forces, not a driving factor. And the policy rate is only loosely linked to real world rates.
I don’t agree on negative interest on reserves. When ERs are near zero, 90% of new base money goes out into cash. If the Fed were to pay a negative 4% rate on ERs, then ERs would be near zero. In that case roughly 90% of ERs would go out into circulation. The Fed most definitely can control the amount of non-interest-bearing currency in circulation.
When the day of interest-bearing currency arrives, currency may not be the unit of account. It might be some fraction of NGDP futures.
Greg, That all sounds wonderful to me.
Morgan, I’ll believe in multiple currencies when I see them in the US.
Jim Glass, Krugman’s much closer to the truth, but 8 trillion is probably way too much.
Mark, Don’t make yourself a slave to the textbook. I often deviate from the text. But I agree with you that texts can be frustrating.
Lee Kelly, Yes, I added an update. I think interest rates have their uses, but I believe we can do business cycle theory w/o them. I agree that to explain saving and investment we need interest rates.
ahhaha, Sorry you are leaving, and sorry to inform you that you are still wrong about Strong. He policy style did not outlive him.
123, How do interest rates help us forecast? I understand that one can infer market forecasts from examining the yield curve, but lots of other things are important as well. But suppose we weren’t trying to infer market forecasts, but rather create our own forecast. How would interest rates help us forecast the future? Does a fall in rates mean growth will be faster, or slower?
24. October 2010 at 07:18
Scott, can we at least agree that when you see multiple currencies you’ll be out of a job?
I’ve now seen Krugman at 6% and $8-10T… of course we’re going to get $100B until 2%.
As such Scott, doesn’t that mean you should start in with suggestions to deflate wages and (home) prices as quickly as possible?
After all, the more these things fall, the more QE you’ll get (to stay at 2%).
Is there something I’m missing here?
24. October 2010 at 13:14
Huge plug for Scott Sumner on Seeking Alpha:
http://seekingalpha.com/article/231848-qe2-revisiting-milton-friedman-s-advice?source=yahoo
The author starts out:
“QE2: Revisiting Milton Friedman’s Advice”
“Economist Scott Sumner makes a good point: “For decades the Fed has steered the economy along a path of two to three percent inflation…”
Unfortunately, below are some examples of how the “Man on the Street” responds to QE2 (Reminds me of a Fisher/Hoenig speech):
“All QE2 does is lead to rewarding greed, excessive risk, punish savers…”
“We just keep growing so that GDP can be positive, ignoring the truth that we are running out of commodities.”
“It’s a balance sheet recession. Until debt and asset prices fall to their ‘natural’ level, no healthy recovery is possible. ”
“Until the structural damage is corrected, and the country returns to sound money with minimal government intervention in the markets, there will be no sustained recovery. Milton Friedman, who at least was an honest man, would admit his mistakes if he were around today.”
24. October 2010 at 13:42
Scott,
You wrote:
“Mark, Don’t make yourself a slave to the textbook. I often deviate from the text. But I agree with you that texts can be frustrating.”
Not exactly my point Scott. My point was that others think textbooks are so important they write their own completely unnecessary versions.
On the other hand Macro is the one area where there really is a need for a good textbook. Thus, although I would normally be extremely critical of anyone writing any textbook, I actually am encouraging you to write one.
24. October 2010 at 14:35
Mishkin = Plosser
http://www.ft.com/cms/s/0/4b6276f8-df95-11df-bed9-00144feabdc0.html
24. October 2010 at 14:49
via JimP
Mishkin wrote:
“This should not be too difficult, because the Fed’s longer-run inflation projections indicate FOMC members already think this rate should be about 2 per cent, or a bit below.”
Jimp this sounds expansionary relative to the current data.
The six month PCE annual rate is 0.4%. The 6 month PCE annual rate excluding energy and food is 1.1%. The 6 month PCE annual rate trimmed mean is 0.7%.
Am I missing something? Isn’t Plosser shooting for price stability. On the other hand isn’t Mishkin suggesting that opportunistic disinflation is a bad idea?
24. October 2010 at 15:08
Mark –
Yes – but a rate target rather than a level target is a promise to deflate – later rather than sooner perhaps – but still – to deflate rather soon.
The problem I think is to convince people that the Fed is actually serious – that they want (if they do) a higher permanent price level.
DeLong says:
http://delong.typepad.com/sdj/2010/10/why-quantitative-easing-needs-to-involve-securities-other-than-government-securities.html
begin quote
Suppose, however, that the nominal interest rate on T-bills is zero and that you cannot alter inflation expectations–cannot commit to keeping your quantitative easing permanent, cannot commit to an exchange rate path, whatever, you cannot do it and inflation expectations are immovable. Then what?
Then, as Paul Krugman says, quantitative easing is working be altering the spread between the short-term safe T-bill rate and the long-term, systemic-risky, beta-risky, default-risky rate. How does it do that? Lloyd Metzler and James Tobin would say that it does so by altering relative asset supplies–by taking duration risk, systemic risk, beta risk, and default premia off of private savers’ books and placing them on the government’s books (and thus on the taxpayers, who are a very different group of people than are private savers). To the extent that quantitative easing thus involves assets whose risk characteristics are very similar–federal funds and two-year T-notes, say–we would not expect even a lot of quantitative easing to have much of an effect on anything.
end quote
I think he is right. The Fed needs to buy risk. I doubt Miskin agrees.
24. October 2010 at 15:19
Or – to put the point in a different way –
as some guy said – the problem is there is just not enough money around.
24. October 2010 at 15:22
JimP wrote:
“Mark –
Yes – but a rate target rather than a level target is a promise to deflate – later rather than sooner perhaps – but still – to deflate rather soon.
The problem I think is to convince people that the Fed is actually serious – that they want (if they do) a higher permanent price level.”
We’ve already deflated with no promise to return to that previous trend. I’ll settle for a percent change target of 2% on our deflated trend. I’d love to get to the previous trendline but realistically it will never happen. There’s just too many inflation hawks in charge (and based on the forthcoming elections, there’s more on the way).
24. October 2010 at 15:37
Mark –
I know. They will never go for a level target. It is depressing.
As Scott keeps saying – if only all this was understood not in terms of inflation (bad bad bad) but rather in terms of growth (good good good). Its just sadism to want what we now have – but somehow we have talked ourselves into it.
What DeLong advocates will of course not happen. The Fed cant buy private assets. DeLong is setting up a straw man there – the only thing that will work is something the Fed cant do.
24. October 2010 at 15:45
JimP,
You wrote:
“What DeLong advocates will of course not happen. The Fed cant buy private assets. DeLong is setting up a straw man there – the only thing that will work is something the Fed cant do.”
It doesn’t matter at all what the Fed buys. It only matters if they buy. Hand me the printing press. I’ll have us hitting the old NGDP trendline in no time flat.
24. October 2010 at 17:42
JimP has linked to it but I think there´s some “desperation” to Mishkin´s arguments. In 3 sequential paragraphs he uses the expression: “the FOMC should announce that this rate would only be modified for sound economic reasons” and in one repeats it as: “by stating its intention not to modify the rate without a clear technical rationale”. He also tries to dispel worries that the Fomc would put too much emphasis on inflation and not enough on growth/employment:”Some commentators have worried that establishing an inflation objective will soon lead to an overemphasis on controlling inflation, and not enough concern about stabilising real economic activity”.
All in all, a very confusing piece. Much better to propose the “all encompasing” NGDP stability! It would cover, simultaneously, both p and y! And forget about “sound economic reasons…
24. October 2010 at 21:22
JimP please explain why 2.5% inflation when the level target is 2% is not a promise to deflate?
To me, level targeting at 2% is HAWK CRACK, it is a final answer when and to what degree do we get tight money.
Whether they will commit to a higher price target is uninteresting… whether they can get Scott, DeKrugman, Congress and everyone else to lay down some game rules and live with them when policy says PAIN, who cares what people agree to during PUMP.
25. October 2010 at 05:01
“A macroeconomics free of the concept of inflation, and a business cycle theory that did not use either inflation or interest rates. Is that slightly less crazy?”
If you used distribution of prices across the entire economy, you could do a sort of Box-Jenkins general equilibrium sort of methodology, and get rid of inflation measures. You would still need interest rates though, to measure expectations.
25. October 2010 at 05:03
@Morgan:
And it seems you step into the dynamic time inconsistency problem in one. 🙁
25. October 2010 at 05:04
@JimP
“What DeLong advocates will of course not happen. The Fed cant buy private assets. DeLong is setting up a straw man there – the only thing that will work is something the Fed cant do.”
The law says they can’t buy private assets, but that doesn’t actually stop them. They do buy private assets. IIRC they own/owned the Red Roof Inn for example.
25. October 2010 at 06:27
Morgan, When we have multiple currencies, won’t we need a priesthood of macroeconomists to explain how they work? For instance, will we have multiple units of account? When you shop in stores, will the price tag on apples be listed in one unit of account, and the price tag on oranges another unit? Won’t that be confusing?
Steve, Thanks for the link–that’s good to see.
Mark, Sorry I misread you. I don’t think I’d be a very good textbook writer. Too idiosyncratic. I plan a post on Mishkin’s text, sometime in November.
JimP and Mark. I agree with Mark that it is better than the previous Mishkin post I read, but I agree with JimP that we really need price level targeting.
Marcus, Yes, and all that timidity by Mishkin makes it less likely that QE will change market expectations in any sort of dramatic way.
Morgan, Because if you don’t make up for past undershoots, in the future your INFLATION targeting won’t have much credibility.
Doc Merlin, What’s been happening to the price of “a computer?” Or a car? Or a TV set?
26. October 2010 at 14:48
‘Doc Merlin, What’s been happening to the price of “a computer?” Or a car? Or a TV set?’
Used car prices are way up, Moore’s law adjusted computer prices are steady or slightly trending up (they should be falling), and hedonically adjusted TV prices have been falling.
Its mostly food and items that have high resale value that have been trending strongly upwards, and housing prices falling have biased the CPI score downwards.
27. October 2010 at 02:37
“How do interest rates help us forecast? I understand that one can infer market forecasts from examining the yield curve, but lots of other things are important as well. But suppose we weren’t trying to infer market forecasts, but rather create our own forecast. How would interest rates help us forecast the future? Does a fall in rates mean growth will be faster, or slower?”
Interest rates contain useful information about money supply expectations that is not available elsewhere. If you have a monetarist model, you can construct your own forecast. For example, 3 month LIBOR started falling in mid-October 2008, this means that money supply is not expected to contract as sharply as before, this means NGDP/RGDP crash will be softer.
27. October 2010 at 15:26
From a traders perspective, a single interest rate by itself is not very useful. But you can use interest rates as a proxy for the money supply/liquidity. Knowing future auction schedules for different central banks a trade position best set. For example, short a currency and be long on equity that is interest rate sensitive, or short one interest rate and go long on another. Forecasting is done using this logic, but given the level of intervention today, its risky.
27. October 2010 at 15:27
From a traders perspective, a single interest rate by itself is not very useful. But you can use interest rates as a proxy for the money supply/liquidity. Knowing future auction schedules for different central banks a trade position can be set. For example, short a currency and be long on equity that is interest rate sensitive, or short one interest rate and go long on another. Forecasting is done using this logic, but given the level of intervention today, its risky.
27. October 2010 at 17:16
Doc Merlin, You don’t understand my point, all that ‘hedonics’ is pure guesswork. Moore’s law has almost nothing to do with how useful a computer is to me. My Word Perfect program from 15 years ago was far more useful to me than the newest Microsoft Word. The first guy in my neighborhood to get a black and white TV derived much more utility that I get from a plasma.
123, I think exactly the opposite–falling interest rates mean the crash will be sharper.
Randall, Why use i-rates as a proxy for M, Why not just look at M?
28. October 2010 at 02:12
Scott,
Stock market does not agree with you – when 3m LIBOR started falling, stock market crash stopped.
You said:
“Why use i-rates as a proxy for M, Why not just look at M?”
M is not interesting, expected M is.
30. October 2010 at 21:46
Scott: characterising interest rates as the rental cost of money has an obvious appeal — it is about money as an asset.
Karl: So I should say interest is the opportunity cost of holding money. What you give up. I suppose that’s not technically a price in the sense of a rate of exchange. Not quite. You are giving up buying whatever goods and services you could with the money. But the longer you do, the greater the chance you will never do so. Hence the (increasing over time) need to be compensated for that. So, when folk talk of the “risk-free cost of capital” as the base interest rate, what they really mean is that element of risk that cannot be eliminated due to human mortality and aging (since aging reduces consumption possibilities too).
For, if there was absolutely no risk (including no aging or mortality), there would be no cost in deferring use of the money.
Other things being equal, the “risk-free cost of capital” should decline as life expectancies increase, be lower in societies with reliable inheritance rights (assuming at least some propensity for wanting your progeny to prosper) and be zero if you are loaning from God (or were God).
1. November 2010 at 00:23
@Scott:
“Randall, Why use i-rates as a proxy for M, Why not just look at M?”
its impossible to measure the real M, especially as more and more liquid assets get used for money, derivatives also have monetary effects but can be impossible to aggregate into useful Ms.
“You don’t understand my point, all that ‘hedonics’ is pure guesswork. ”
I agree, its what makes CPI so incredibly useless.
“Bill, I am a pragmatist, and prefer to model monetary systems that actually exist, or have existed, in the real world. None of those involve interest-bearing currency”
Nowadays, currency bears interest. Also, substitutes for currency also bear interest.
4. November 2010 at 17:15
123, Most of the time during 2008 stocks and interest rates moved in the same direction–lower. See my new post on Fed announcements.
Lorenzo, I agree that interest rates are the rental cost of money.
Doc Merlin, I don’t think M or i are particularly interesting.
5. November 2010 at 03:43
If expected PY is interesting, then expected M and expected V are also interesting. And i is very sensitive to expected M and expected V.
7. November 2010 at 04:58
123, Again the relationship between i and expected NGDP is ambiguous. The liquidity effect cuts one way, and the Fisher effect the other.
Why isn’t something like stock prices a better indicator? Stock prices should rise if expected future NGDP rises?
I’m not saying interest rates contain no information, the slope of the yield curve is interesting. I just don’t find them to be particularly interesting. I prefer asset prices.
9. November 2010 at 07:23
Yes, the relationship is ambiguous, but it is often possible to determine the cause.
Stock prices are also very useful, but they have the same ambiguity too, as they can be decomposed into required return on stocks (which is an interest rate for which both liquidity effect and Fisher effect apply), and forecasted nominal profits.