Nothing new here. Think of this as a teaching post. Partly for the people of the concrete steppes, but for others too. I'm just going to talk about different ways that expectations can affect what happens. The main distinction is between multiple equilibria and unique equilibrium.
1. Multiple Equilibria
Here is one simple model of multiple equilibria: Y=E(Y)
Let Y be something that people do. Let E(Y) be people's expectations of what people do. The simple model Y=E(Y) says that people do whatever they expect other people to do. Expectations are totally self-fulfilling. People can expect anything they like, and it happens, just as they expect. There is a very large number (a continuum) of equilibria. Anything whatsoever can be an equilibrium. Or almost anything, if Y=E(Y) is only true over some range.
Here is another simple model of multiple equilibria, but where not everything is an equilibrium. Y=[E(Y)]2 + X, where X is some exogenous variable. This model has two equilibria. For any given level of X, there are two different levels of Y at which expectations will be self-fulfilling. But if people expect some other level of Y, their expectations will be falsified by their actions. And if X changes, both equilibria will change too.
It is easy to construct a model with three, or four, or any number of equilibria. Depending on how people form their expectations, and how they adjust their expectations if they turn out to be false, some of those equilibria may be unstable, and others stable.
Examples of multiple equilibria are all around us.
I use the word "cat" to mean cat, because I expect other people
around me to to do the same. (But when I'm talking to some other people
around me, I use the word "chat" to mean cat, because I expect those
people to do the same). There is no reason why we couldn't have a
different equilibrium in which we all used "cat" to mean dog, and "dog"
to mean cat. Almost any word could mean cat.
Another example is
which side of the road to drive on. Here, Y can only take on two values:
"left" or "right". If you expect everybody else to drive left, you
drive left, and if you expect everybody else to drive right, you drive
right.
Most people obey the law and follow the leader because they expect everybody else to obey that same law and follow that same leader. If everybody thought the law was a different law, or the leader a different leader, they would obey that different law and follow that different leader. The law is whatever people expect to be the law. The leader is whoever people expect to be the leader. Many different things (though perhaps not anything) could be the law, if people thought it was the law. Many different people (though perhaps not anyone) could be the leader, if people thought that person was the leader. The constitution, which is supposed to tell us who gets to make the law and who gets to be leader, wouldn't be the constitution if people didn't think it was.
2. Unique Equilibrium.
Here is a simple model with expectations and a unique equilibrium. Y=E(X). It is not X that determines what people choose to do; it is people's expectations of X that determines what people choose to do. It is not rain, but our expectations of rain, that determine whether we carry an umbrella.
Here is a slightly more complicated model. Y = 0.5E(Y) + 0.5X. What people choose to do depends partly on some exogenous variable X, and partly on what they expect other people to do. But there is only one level of Y, given X, at which people's actions will confirm their expectations of others' actions. Most static economic models are like that.
Now lets complicate that model slightly, by introducing time. Y(t) = 0.5E(Y(t+1)) + 0.5X(t). What people do today depends partly on X today, and partly on what people expect people to do tomorrow. Most dynamic economic models are like that. And you need to say something about how people form their expectations in order to solve those models.
One assumption is rational expectations. People's expectations are consistent with the model. Taking that same equation and leading it forward one period gives you Y(t+1) = 0.5E(Y(t+2)) + 0.5X(t+1). If people expect that that is how people will choose tomorrow, then E(Y(t+1)) = 0.5E(E(Y(t+2))) + 0.5E(X(t+1)). We can substitute this solution for E(Y(t+1)) into the first equation, repeat again and again, and get the equilibrium as:
Y(t) = 0.5X(t) + 0.52E(X(t+1)) + 0.53E(X(t+2)) + etc.
[Somebody please tell me politely if I've screwed up the math, as I usually do.]
The solutions to almost all modern macroeconomic models look like that. If X is some policy variable, then it is not just current policy that matters, but people's expectations of the whole path of future policy.
If you make a different assumption about expectations you will get a different solution. But if people think that policy affects what people do, and also think that expectations affect what people do, you will still get some sort of solution which looks something like that.
In that example, I used "0.5" as a numerical parameter value for both E(Y(t+1)) and X(t). That means that Y(t) is determined 50-50 by current policy and by expected future policy. If I had originally measured time in years, and then switched to measuring time in 6-month periods, I would need to change those parameter values, and make the model (roughly) Y(t) = 0.75E(Y(t+1)) + 0.25X(t). And the solution would now become:
Y(t) = 0.25X(t) + 0.752E(X(t+1)) + 0.753E(X(t+2)) + etc.
[Damn. i've screwed up the math there, haven't I. I need more coffee, or some help.]
Now Y(t) is only 25% determined by current policy, and 75% determined by expectations of future policy.
As we shorten the time-period still further, we approach a solution in which Y(t) is 99.99% determined by expectations of future policy.
3. Counterfactual conditional expectations matter too.
I do not break the speed limit. (OK, just assume I don't.) I do not expect the police to give me a speeding ticket. But I do expect the police would give me a ticket if I did speed. Which is why I don't speed. So in equilibrium I do not speed. And I don't expect the police to give me a tcket.
What people do in equilibrium is in part determined by people's counterfactual conditional expectations of what other people would do away from the equilibrium.
4. Monetary policy at the ZLB.
New Keynesian macroeconomic models have solutions that look like:
Y(t) = 0.25X(t) + 0.752E(X(t+1)) + 0.753E(X(t+2)) + etc.
Except:
4.1 For quarterly data, it's a smaller number than 0.25 (and so a bigger number than 0.75) and so expected future policy matters a lot more than current policy;
4.2 The policy variable X(t) is interpreted as a nominal interest rate, but the nominal interest rate is not exogenous. It is determined by a feedback rule so that it depends on Y(t). This means that people's expectations of policy can be described [I should have said "and must be described"] as beliefs about the parameters in that feedback rule, rather than expectations about the nominal rate of interest itself. In all New Keynesian models, if the central bank announced and stuck to a time-path for X(t) that was not a feedback rule, (and so did not depend on Y(t)), the model would have multiple equilibria, with nearly all of those time-paths for Y(t) leading to an explosive or implosive path for the inflation rate. (If expected inflation were to increase, that would reduce the real interest rate for any given nominal interest rate, which would increase demand for goods, which would increase inflation, which would increase expected inflation, and so on.) What prevents those explosive or implosive paths happening are people's expectations about what the central bank would do if one of those paths were to happen (it would increase nominal interest rates by enough to stop it happening). This means that counterfactual conditional expectations matter too in New Keynesian macroeconomic models.
Even if you think about monetary policy as New Keynesians do, as setting a counterfactual conditional feedback rule for the nominal interest rate, this does not mean that monetary policy cannot be loosened at the ZLB. If people expect that at some future time, however distant, the economy will escape the ZLB and the central bank will want to raise interest rates to prevent the inflation rate rising too much, monetary policy can still be loosened today. The central bank can commit to delaying the time at which it would otherwise increase the nominal interest rate. It could commit to keeping interest rates "too low for too long". A better way to communicate that commitment might be to commit to a higher price level or NGDP level in the future feedback rule for monetary policy. And it might be better still to stop talking about monetary policy in terms of interest rates, and describe the monetary feedback rule in terms of a setting for some asset price as a function of an NGDP target.
I don't have great hopes for this post persuading people to change their perspective on expectations. But one can only try.
Worthwhile Canadian Initiative 
I’m glad you started with multiple equilibria. The existence of a relationship between current income and desired expenditure such that there can be many equilibrium levels of output with the same “fundamentals” (including the same interest rate) is one of the key insights of the Keynesian revolution, but it tends to disappear in the versions of Keynesian economics that get taught today.
On monetary policy, obviously we have two questions. First, is it in fact possible for the current decisionmakers at a central bank to change expectations about what the central bank will do in the future? I would say there are good reasons to think No. Central banks cannot make legally binding commitments about future policy, today’s officials will not be in office forever, there is no guarantee that the ZLB will cease to bind over the relevant period, the actual expectations-formation process is largely backward looking, etc. There is a close parallel between claims made for expansionary expectations management today, and the claims made for disinflationary expectations management by Monetarists in the late 1970s. In that episode, despite the promises of Friedman et al., the “credible commitment” of the central banks to slower money growth did not noticeably reduce the output costs of disinflation. Not clear why central bank promises should work better today.
Second, even supposing that it is possible for the central bank to raise inflation expectations, will this in fact raise desired current expenditure? It is true that that real long-term interest rates will be lower if borrowers expect higher inflation over the life of the loan. Structures and other long-lived assets will also have a higher present value relative to financial assets. (But so will land…) These are the expenditure-raising channel. But to know if they will dominate, we need to know more about the costs of inflation. Weirdly, despite controlling inflation being the central goal of macropolicy for much of recent decades, there doesn’t seem to be a good account of why inflation is so costly. But given that private actors clearly do find it costly, it is quite possible that the expectation of higher inflation will lead businesses to reduce investment, households to reduce purchases of durables, etc. As it happens the 1970s in the US were a period of very high fixed investment; but it would be easy to point to inflationary episodes elsewhere where investment stayed low.
So as far as I’m afraid your conclusion is right: You haven’t changed my mind, at least, that the expectations channel is not a realistic transmission mechanism for monetary policy when the central bank cannot move current interest rates.
How did your expectations of failure shape your execution of this post?
More ‘concretely’ I think the expected length of the ZLB would be pretty critical in the effectiveness of monetary policy in the NK framework. It affects both the discounting of inflation costs (if the ZLB is expected to last 10 or 20 years as in Japan that may even be longer than my expected asset holding period) and the credibility of the CB’s commitment, the longer the ZLB lasts the less credible any commitment made by current officials will be.
JW: I agree with you on multiple equilibria. I think they are really important. And not just in the (narrowly-defined) “economic bits” of society. Everything we think of as “civilisation”, or “society” is one equilibrium out of many. And civilisation advances (and regresses) in large part by jumping from one equilibrium to another. And it’s weird for me to see people who think they are “Keynesian” being sometimes the most likely to give a concrete steps theory of fiscal vs monetary policy, and make dismissive remarks about self-fullfilling expectations.
On your second point, you need to remember the real income channel too. If the Short Run Phillips Curve says that increasing AD will see both higher inflation and higher real income growth, I would say the effects on demand of higher expected real income growth would swamp the effects on demand of any real costs of inflation. Keynes, man, Keynes!
It doesn’t need to be some cast iron legally binding commitment. As long as it changes expectations a bit, it will help. And as we approach a world of multiple equilibria, where the forces keeping the economy in one bad equilibrium get weaker and weaker, we need less and less to escape. At the limit, a new Schelling focal point will do the job. Cheap talk, or a sunspot, is enough. Think how easy it is for the government to get us to change our clocks every spring and fall.
OGT: “How did your expectations of failure shape your execution of this post?”
If I had expected certain 100% failure with every reader, I would probably not have written it. If I had been more confident of success, I would have written it sooner.
Yes. It would have been better if monetary policy had been more aggressive a lot sooner. But a P or NGDP level path target helps offset the time we expect to spend at the ZLB. Plus, if the policy is implemented, we escape the ZLB much more quickly.
Nick
Suppose the inflation expectations that do form are such that people only expect commodity inflation but do not expect their wages to go up. What does the Philips curve look like in such a scenario?
Heterogeneity of the composition of inflation expectations may itself be a great source of the existence of multiple equilibria. And real income expectations are not invariant to this heterogeneity. There are as many Philips curves as there are equilibria of inflation expectations.
I have a question.
A central bank that controls the money supply can successfully target a wide range of nominal variables (NGDP, PL, wage levels etc) simply by varying the money supply and seeing how if affects their chosen variable even without the use of expectations. Setting expectations however makes it possible to achieve the target without actually varying he money supply so much.
If I am understanding this correctly then in terms of using these nominal goals to drive real variables (RDDP, employment levels) then expectations setting is not just an easy way to achieve the nominal target but a way to make the nominal variables have a closer correlation to real ones. For example if the CB sets an NGDP target for the next period then the actual mix of inflation and RGDP growth that occurs will depend upon how effectively it communicated the target and hence controls behavior now that will influence RGDP in the future.
So my question is: Given that setting nominal targets is just a way of controlling expectations and assuming the CB knows what kind of expectations today are appropriate for achieving an optimal future why doesn’t the CB target expectations and use the various nominal variables it control to achieve this, rather than targeting the variables that may have indeterminate effects on real variables ?
Ritwik: the short answer to your question is: I don’t know. Given our ignorance about the short run Phillips curve, our ability to answer more complicated questions like that is very limited.
Ron: “Given that setting nominal targets is just a way of controlling expectations and assuming the CB knows what kind of expectations today are appropriate for achieving an optimal future why doesn’t the CB target expectations and use the various nominal variables it control to achieve this, rather than targeting the variables that may have indeterminate effects on real variables ?”
Well, inflation targeting central banks certainly do part of this. The Bank of Canada spends a lot of time repeating that it is targeting 2% inflation, and that people really should expect 2% inflation, and that it will do whatever it takes to bring inflation back down to or back up to 2% over the “medium term”. But yes, why do they still talk about adjusting an interest rate as the transmission mechanism via which they will hit their target? That would take a long essay in the history of thought, plus perhaps the influence of commercial bankers’ way of thinking. Commercial bankers think in terms of interest rates. And so they tend to think that the Bank of Canada is a bank. Which it isn’t, really.
A central bank that controls the money supply
We don’t live in a world where central banks control the money supply, tho.
“Nevertheless, circumstances can develop in which even a large increase in the quantity of money may exert a comparatively small influence on the rate of interest… But if not, then, if we are to control the activity of the economic system by changing the quantity of money, it is important that opinions should differ.”
“[A liquidity trap situation is one in which] a moderate increase in the quantity of money may exert an inadequate influence over the long-term rate of interest.”
“A change of the quantity of money… is already within the power of most governments by open-market policy or analogous measures.”
I wonder what world Keynes was living in?
W. Peden,
Keynes lived in a world where central banks did control the money supply. (Or at least where that was a reasonable first approximation.) There’s a great discussion of this by Leijonhufvud in The Wicksellian Heritage.
Economic laws are not like physical laws; they are historically contingent and change as economic institutions evolve.
JW Mason,
Yes, and so do we. Private money creation was as much a feature of Keynes’s world as our own. Or as Keynes puts it in the Treatise- “… we thus have side by side State money or money proper and bank money.”
We just disagree. For much of the 20th century, money creation by banks was effectively limited by binding reserve requirements and related regulations, so that there was a stable relationship between high-powered money and credit money. Today, it is not and there isn’t.
If you want to see where I’m coming from read the Leijonhufvud piece linked in my previous comment. I’m not going to try to improve on it here.
It seems a truism of economics that it is expectations about the future that determine decisions today. Business will invest today based on expectations about future prices and sales in the period goods starting production now will be sold.
The theory behind the use of nominal targets seems to be as follows: CBs use nominal targets to influence these expectations about the future that drive investment. It turns out that setting expectations on what nominal targets the CB is going to fulfill makes it easier for them to achieve them (ie: they can reach the target without changing the money supply so much as without the use of expectations). Further: The use of expectations to achieve a nominal target actually helps change the kind expectations that drive investment. A credible commitment to meet an NGDP tomorrow will drive investment today as it increases confidence about sales and prices tomorrow.
My concern with the above is that it only works as long as the nominal targets work on expectations in the way that the CB hopes it will. If an NGDP target doesn’t inspire more investment it may just raise inflation fears and cause people to swap money (that they otherwise would have invested) for gold and other safe assets and hence reduce RGDP compared to the situation with no NGDP target. In this situation an inflation target would have been better even if it led to lower NGDP in the short run.
So given that a CB can never really know the real effects of a nominal target should they not target expectations of real variables? For example supposing that there is a correlation between expectations of RGDP growth and actual RGDP growth – could not the CB target an expectation of 3% RGDP growth by manipulating the various nominal variables it has available (NGDP, inflation and interest rates etc)? It could use a RGDP futures market or a just a survey of business to ascertain what RGDP expectations actually were.
J. W. Mason,
That a causal relationship has become unpredictable does not mean that it has ceased to exist. A monopoly supplier of base money can still hit any nominal target it likes over the medium-term, provided it’s willing and legally able to utterly destroy the economy. Hitting a target for the price of broad money (the price level) is no easier than controlling the quantity of broad money. Indeed, central banks only control the former insofar as they can adjust the latter.
Of course, a hardcore endogenous money type can deny that the central bank can control either, and I don’t think either of us is going to provide the other with an argument tonight that will change the other’s mind. Nevertheless, I do think that it’s interesting and useful to consider one’s position on these matters and also to consider the position of Keynes. In particular, it’s worthwhile to note that Keynes’s liquidity trap is very different from that of New Keynesians or Post-Keynesians. I’m not sure if anyone still believes in the liquidity trap in the GT, though monetarists and Keynesians and Robertsonians will all disagree among each other on WHY it’s wrong!
I like the liquidity trap in the GT. I’ve written a bunch of blog posts defending it, in fact.
I agree with you that neither of us will change the other’s mind!
Nick,
Basic stuff:
I find it concretely obvious that when a Canadian mortgage borrower takes out a variable rate mortgage, in making that decision he/she typically ponders the risk of rates rising in the future and the timing of that risk. And that risk is directly affected by the expected timing of Bank of Canada interest rate policy. And that is a consideration that holds equally true at the zero bound, or at any policy rate of interest. Are you suggesting there are thinking people who don’t understand or consider that?
And when people buy a stock or a bond, they implicitly or explicitly discount the value of expected future cash flows to the present. What is the general problem about expectations here? At its broadest level of generality, which also applies to monetary policy, it seems to be a trivial issue. Are there people who actually don’t think about translating potential future consequences to present action? What am I missing about the premise of the importance of this basic concept of expectation as it applies to monetary policy, because some people certainly seem to make a huge deal about it?
The problem being that the central bank cannot commit “to delaying the time at which it would otherwise increase the nominal interest rate. It could commit to keeping interest rates “too low for too long”. ”
We don’t even know if Bernanke will be re-elected. We don’t know who will be president. We don’t know what cutting edge macro research will be in the future.
The very fact that we are talking about changing monetary policy today means that we can also change it to tomorrow.
So it is very difficult for me to believe that the CB will keep rates too low in the future merely because it would be in the interest of the CB to make such a promise today. In the future, we will be beyond the ZLB. If the ZLB happens once every 60 years, it is far too easy for the CB to forget those promises and start fighting inflation in the future.
And IIRC, economists are blaming the CB for keeping rates too low in the 1950s and 1960s, even though we were in a ZLB in the 1930s and 1940s. I don’t believe we are looking back at that period today, saying, well, they needed to keep rates low because investors assumed rates would be low for an extended period of time and this got us out of the Great Depression. What can the CB do today that would force it to adopt a certain monetary stance two decades from now?
The above should read “re-appointed” instead of “re-elected” 🙂 But you get my point, I hope.
Nick, speaking for at least some of us out here in the Concrete Steppes I would just like to report that many of us understand how these a priori models based on rational expectations are supposed to work. We just don’t believe that the world we live in is correctly represented by the models. (And I’m not even talking about the fact that some of us don’t believe that people have rational expectations.)
Here’s a simple model describing a class of possible worlds: in this world there are two people: Master and Servant. Servant has a fixed psychological disposition to form comprehensive expectations of the future based on the statements of Master about the future. Servant also has a fixed behavioral habit of acting in the way Servant’s comprehensive expectations of the future say Servant will act. Master also has a fixed psychological disposition to form comprehensive expectations of the future based on the statements of Master about the future. And Master also has a fixed behavioral habit of acting in the way Master’s comprehensive expectations of the future say Master will act.
There are multiple outcomes consistent with these assumptions:
1. Master says, “Five minutes from now, you and I will be doing the Tango”; and indeed, five minutes later Master and Servant are doing he Tango.
2. Master says, “Five minutes from now, you and I will be making mud pies”; and indeed, five minutes later Master and Servant are making mud pies.
Etc. etc.
Now here are my neighbors, Bob and Carol. Five minutes ago, Bob said. “Five minutes from now, you and I will be doing the Tango.” And yet, Bob and Carol are not doing the Tango. What went wrong???!
Nothing went wrong. It’s just that the actual circumstance of the household in which Bob and Carol live is not adequately represented by the logically possible model of the Master and Servant worlds. That a priori conceivable model is not a correct description of the actual household in the actual world.
And there is no suggestion here that Bob and Carol don’t have expectations that are rational through and through. Instead, they are simply lacking the kinds of dispositions and habits described in the account of the Mater and Servant worlds.
I understand the forward guidance story about people whose expectations of future interest rates are determined in part by their expectations of various kinds of actions the central bank might take in response to various kinds of changes in the inflation rate; and whose economic decisions now are based on their expectations of future interest rates. I just don’t think the universe of people whom that story accurately represents is, in a US economy of 300 million people, a very significant number of people.
The challenge from the Concrete Steps is an empirical challenge about the actual behavioral facts, psychological facts and institutional facts of the actual economy we live in. That’s why our steppes are called “concrete”. It’s not a challenge about the bare theoretical intelligibility of descriptions of various kinds of economically possible worlds.
Good post about an underappreciated issue!
The very fact that we are talking about changing monetary policy today means that we can also change it to tomorrow.
Exactly. The fact — if it is a fact — that central bank A’s commitment to some rule produces economic effect X, does not mean that central bank B could commit to a similar rule and get a similar outcome. The fact that A is committed to the rule means precisely that does not have a choice about it. As soon as we say that central bank B is choosing whether to adopt the same rule as A, we have ruled out the possibility of it being committed in the way A is, because that commitment is defined precisely by not having (or not being believed to have) a choice.
(This goes back to the Chuck Norris posts here a few months ago.)
And IIRC, economists are blaming the CB for keeping rates too low in the 1950s and 1960s, even though we were in a ZLB in the 1930s and 1940s.
This is a really good point. Looking retrospectively, nobody explains economic outcomes in terms of the monetary policy 10 or 20 years later. Nobody. And of course to define “rational expectations” in a way that says people in 1935 knew the true expected value of the short interest rate a decade after World War II (which presumably implies that they knew there would be a war, what its outcome would be, etc.) is to define the term in a way that’s lost all contact with any recognizable human reason.
Dan Kervick’s comment is good too.
If the Short Run Phillips Curve says that increasing AD will see both higher inflation and higher real income growth, I would say the effects on demand of higher expected real income growth would swamp the effects on demand of any real costs of inflation.
Sure, if the central bank is believed to be able to stabilize real output, then that expectation will itself help stabilize real output. This is undoubtedly one of the reasons that the quantitatively trivial operations in monetary policy can be effective at maintaining full employment. But it’s a different story when you’re no longer talking about stabilizing the existing level of output, but closing a large output gap – especially when the central bank has been unsuccessfully trying to close that gap for years. If the current effectiveness of policy depends on the beliefs about its future effectiveness, seems to me we’re back in multiple equilibria land. If we can only have jam today if we have jam tomorrow, then we always have jam or we never do.
Monetary policy is better for preventing depressions than for ending them.
It’s even worse, I think, actually. If future expansionary policy would raise output today, and if output today nonetheless remains depressed, then rational agents should adjust their priors about the likelihood of future expansionary policy downward. So the longer the depression goes on, the harder it gets for the expectations channel to work.
Sorry — one more comment. I think some of the confusion here comes from importing model-logic into analysis of actual economies.
If I’m making a macro model, I’ll write down some structural equations (say, an IS relationship linking the output to interest rates and a Phillips Curve linking inflation to the output gap), and then write a monetary policy rule linking interest rates to inflation and/or output. Once I’ve written those, and specified the distribution of shocks my economy is subject to, I can then ask what the expected value at time t is of some variable at time t+n. I can try this out with different monetary policy rules, and each will imply a different set of expectations.
It’s natural to imagine actual monetary policy the same way — just like the economist can choose what monetary policy rule to put in the model, the central banker can choose what monetary policy rule to follow, and in either case, people being rational, an appropriate set of expectations will follow. But in the case of the real economy this is logically incoherent. Either people have true knowledge of the distribution of future monetary policy, in which case that distribution already exists and is not available as a choice for the monetary authorities. Or else the authority really is free to choose a rule, in which case there is no rational basis for knowledge of future policy and expectations must be nonrational.
“If the ZLB happens once every 60 years, it is far too easy for the CB to forget those promises and start fighting inflation in the future.”
It could also forget the promise to fight inflation when the target is reached. The commitment problem is two sided, and people will also worry about overshooting the target.
Nick
I agree, my question was a complicated one. But my positive argument is that the expectations formations process might itself depend on which concrete steppe transmission mechanism is dominant. For example, if monetary easing is being funnelled into commodity speculation, people will not expect their wages to go up, while commodity prices will. This will probably be stagflationary. If it is being funnelled into mortgages, we should expect new construction or at least improved balance sheets due to mortgage refi. If it is a generic interest rate cut, we should expect more consumption and investment. But we may even expect less investment if we have a model where there are two types of businesses – those that need to borrow to invest (entrepreneurs) and those that don’t (incumbents), and the relative advantage of incumbents is higher when rates are higher, so that aggregate investment is higher when rates are higher. And so on and so forth.
Essentially, my points are :
1) The AS/Philips curve is not invariant to the monetary regime
2) De-aggregating and then re-aggregating might be helpful in understanding why expectations can be/ are being frustrated despite ‘mattering’.
I agree that monetary policy can be loosened at the ZLB. My contention is that the question of ‘which’ monetary policy is always important, but especially important at the ZLB. And that the effectiveness of expectations policy is not invariant to the concrete steppes that form the substrate on the basis of which these expectations will be formed.
I would like to endorse and follow up on Ritwik’s statement:
… my positive argument is that the expectations formations process might itself depend on which concrete steppe transmission mechanism is dominant.
Consider again my example of Master and Servant, but change the assumption that Servant has a fixed psychological disposition to form comprehensive expectations of the future based on the statements of Master about the future. Assume instead that Servant has a variety of conditional dispositions to form expectations of the future based on statements of Master about the future when those statements are accompanied by certain other actions that Servant believes are causally related to the future event Master describes.
So for example, suppose every year Master plows either the Front Field or the Back Field. The Front Field is perfect for raising pumpkins but won’t yield grapes. The Back Field is perfect for producing grapes but won’t yield pumpkins. Servant knows these facts. Servant might have the following conditional dispositions:
1. If Master plows the Front Field and says “We will be harvesting pumpkins this year,” Servant will form the expectation that they will be harvesting pumpkins.
2. If Master plows the Back Field and says “We will be harvesting grapes this year,” Servant will form the expectation that they will be harvesting grapes.
3. If Master plows the Front Field and says “We will be harvesting grapes this year,” Servant will not form the expectation that they will be harvesting grapes.
4. If Master plows the Back Field and says “We will be harvesting pumpkins this year,” Servant will not form the expectation that they will be harvesting pumpkins.
I think this is relevant to a lot of the discussion about central banks and expectations. My hypothesis would be that few people form their expectations about future variables – inflation, nominal spending, the unemployment rate, etc. – on the basis of statements about central bank expectations alone. But a somewhat larger number of people might form their expectations about those variables from central bank statements that are coupled with other central bank actions that those people believe to be causally connected, in ways that operate independently of expectations-related influences, with the outcome the central bank says it expects.
For a given observer X of central bank statements and actions, it could be that if the central banker says “We expect higher inflation this year”, and that statement is accompanied by a helicopter drop, X will expect higher inflation. But if the statement is not accompanied by a helicopter drop, X will not expect higher inflation.
Now suppose the central banker accompanies the statement with, not a helicopter drop, but a round of something the central banker calls “quantitative easing”. How will X respond? It might depend on what X thinks the term “quantitative easing” refers to. If X thinks it refers to a helicopter drop, then X might form new expectations in the same way as if the central banker had notified the public of a helicopter drop. But suppose X believes that the term “quantitative easing” refers to the purchase of limited types of financial assets, and understands that when it occurs, the financial assets that are injected into the economy in the form of money are offset by the purchased financial assets that are removed from the economy. The psychological response is unlikely to be the same. What the response will be depends on what other beliefs X has about the ways in which different variables in the economy are and are not connected.
And then, of course, in our actual economy, there are a large number of people who, unlike X, pay little or no attention to central bank statements and actions in any way.
Here is my 3 different ways expectations can matter;
Workers, their income and spending/saving habits;
Do people think that people actually expect higher future incomes to go along with their expectations of higher inflation? Is the income channel driven by the inflation expectations channel? I can tell you that most people working for someone else, people who rely on a paycheck from someone (most people) do NOT believe that their paycheck will rise in tandem with inflation. They feel poorer as inflation expectations rise (and they are in real terms quite often) and they will NOT spend more today because they fear high future prices.
Now the owners of a business and their hiring decisions;
Do they behave in a way that induces them to spend/invest as their expectations of future inflation increases? Doubtful. If they think future costs will be higher, they will likely cut present costs, and it looks as if the story of the last few years bears that out.
The entrepreneurs ;
I have come to the conclusion that what these inflation expectations economists really believe is that they see the drivers of our economies as those that take the big risks (true to a degree for sure), those that shoot big….. and they see our economies fate as tied to these “benevolent dictators” so to speak. We must rely on these extremely successful entrepreneurs to do the right thing and protect their investments and not get in their way. Its these guys’ inflation expectations that drive things because these guys are real sensitive to the erosion of inflation on their investments. If they see inflation in the future they will spend now to get the best deal they can.
This may be how a small segment of the population sees things and we will see if rising inflation expectations gives them a sense of urgency to do something productive now.
As a “concrete steppes” guy I dont think monetary policy is completely ineffective, I just think its clumsy, fraught with very bad side affects and too top down. I see it as actually more top down than fiscal policy, which interestingly is the criticism of fiscal policy by the monetarists. Just a different view of things? Or is there an actual definition of “top down” that can be discerned ?
JW Mason,
One additional interesting thing about Keynes’s liquidity trap is that, unlike the New Keynesian and Post-Keynesian traps, it’s immune to Tim Congdon’s ideal solution* to our current problems, which is to expand broad money by the government selling its debt to commercial banks rather than the public (“debt-market operations” as opposed to OMOs). However, like me, Congdon is of the opinion that Friedman gave us reasons to dismiss such dangers in 1956.
* Congdon is a monetarist, but he thinks that expanding central bank asset sheets is needlessly controversial and risky.
Dan K.,
While I’m totally with the spirit of your comments, I think that having just two agents obscures one of the main points, which is the role of the central bank in solving the coordination problem among private actors by providing a focal point for expectations.
On the inflation expectations channel, the important thing is not that households desire higher current consumption when inflation expectations rise, but that businesses and asset-woners desire more physical assets as opposed to financial assets. Production of physical assets requires current labor and other inputs, and thus raises incomes. I’m skeptical about this too, but we need to be clear what story we are criticizing.
(Also, to be clear, in modern economies higher (but still moderate) inflation is almost always associated with higher real wages, not lower.)
Also, this statement makes no sense to me at all:
If they think future costs will be higher, they will likely cut present costs, and it looks as if the story of the last few years bears that out.
Are you saying that in a deflationary environment businesses don’t care about their costs? And how do “the last few years” support this? Inflation expectations have been very low. In any case, inflation means both costs and sale prices will be higher, so margins will be the same. The important thing is that financial commitments taken on now will be less burdensome.
Again, it actually is a historical fact that the period of maximum inflation in the postwar US was also the period with the highest levels of business fixed investment.
I don’t think the expectations channel works — I don’t think that the central bank can change people’s beliefs in that way, for reasons similar to yours. But if we actually did experience a period of elevated inflation it seems pretty likely that output, employment and real incomes would be higher as a result.
Oh, sorry, second part of my comment was in response to Gizzard. (Somehow I thought his/her comment was also from Dan K.)
All Ill say is that most owners of businesses really dont give a crap about what some monetary expert says the “inflation expectations” level is or might be. Most guys running businesses think the costs of their future input materials and health insurance premiums are going to rise. THAT is inflation to them. Not some number released by the fed. And as a result of not seeing any more customers coming through the door to meet those costs they are cutting. That is the reality of how businesses react to higher future costs without the expectations of higher future sales.
I think most businesses know intuitively that if every businesses costs are going up they will NOT pay more in salaries as a response, since salaries are one of the costs they are trying to cut.
I dont think periods of inflation lead to high output, employment and incomes, I think periods of high income, employment and output will lead to inflation eventually.
W Peden
1) How are you differentiating between PK, NK and Keynes’s liquidity traps?
2) In Congdon’s solution, the govt should expand its debt to commercial banks and do what with the proceeds? Does he want a helicopter drop of bonds on banks?
Thanks JW Mason. I don’t think anything essential would be changed if I used a three-person model with two servants and a “ringmaster” who doesn’t participate in the economy, but who only makes statements that set the servants’ expectations because both of the servants have have dispositions to form expectations about their own and the other servant’s behavior based on the ringmaster’s statements about what that behavior will be. Do you think it would make a difference?
Also, in the case of the Fed, wouldn’t we want to model them as a participant of some kind in the actual economy, and not just an expectations-setter? After all, they are buying and selling assets of the same kind as many other participants in the private financial sector.
Ritwik,
I’d say that, approximately-
1. Keynes’s liquidity trap is a situation where an increase in the quantity of broad money cannot reduce the yield on long-term government bonds.
2. The Post-Keynesian liquidity trap is when short-term interest rates cannot be cut any further.
3. The New Keynesian liquidity trap is the ZLB.
The Post-Keynesian liquidity trap is superficially similar to Keynes’s, but there are two important differences: (i) the shift from focusing on long-term to short-term interest rates; and particularly (ii) the move from exogenous money (Keynes) to endogenous money (every(?) Post-Keynesian). Nicholas Kaldor (1982) is very good on the difference between the two and very critical of Keynes on this issue.
As for Tim Congdon: the government just spends as it did previously. So instead of financing public borrowing through debt sales to the non-bank public, the government borrows new money from commercial banks; instead of banks being the first private sector recipients of the new money, suppliers to the government would receive the new money. It’s a change in the funding of expenditure, rather in the level of expenditure. To illustrate with an identity-
Broad money growth = ((A – B) + C) + D
Where A is public sector borrowing, B is debt sales to the non-bank public, C is bank lending to the private sector, and D is the overseas influence on the money supply. Congdon’s policy is to shift borrowing from B to A, thereby increasing broad money growth.
I think he understands that it isn’t really practical in current circumstances, because of the current fashion for independent central banks.
the shift from focusing on long-term to short-term interest rates
Why do you see this as progress? I would think the case that long-term rates matter more for output and employment than short-term rates do, is pretty strong.
JW Mason,
I don’t think it’s progress, at least insofar as Keynesian analysis goes. Short-term rates are interesting for their effects on private lending, but Keynes was right (given his assumptions) to focus on the long-term rate, which is the important rate for large capital projects and therefore (as you say) output and employment.
I prefer Keynes to the Post-Keynesians, the New Keynesians to Keynes, and the monetarists to the New Keynesians (New Keynesianism is mostly monetarism without an appropriate focus on money and a lot of dodgy ratex stuff).
JW: Good find on Leijonhufvud on Wicksell. I can’t remember if I have read it before, or just something similar. It’s funny how we are all still debating Ricardo vs Tooke!
Everyone: there’s something very puzzling about this government commitment thing.
There are theses things called government bonds, where the government makes a commitment to pay a certain amount to whoever owns them, sometimes 30 years in the future. Some of those bonds are even supposed to be indexed to inflation!!! I can’t see any reason why anybody would believe that commitment. Would it actually be in the government’s interest to do what it said it would do, 30 years previously? That would be most unlikely. After all, it will be a totally different government 30 years from now, and it won’t care what a government 30 years previously had said it wanted it do do. Those government “bonds” must be worthless, right?
Nick Rowe,
A very interesting analogy. The UK Treasury issued a 51 year bond last year…
Nick, it seems to me that even 30 years from now the government will have an interest in making good on 30-year old commitments, to preserve its reputation as a borrower and preserve the value of similar commitments it will probably want to make at that time. Also, in the US there is actually a passage in the 14th Amendment to the Constitution that says the validity of the public debt of the United States shall not be questioned. One reading is that even Congress is not permitted to undo the debt obligations that it, itself, has made.
Nick, it seems to me that even 30 years from now the government will have an interest in making good on 30-year old commitments, to preserve its reputation as a borrower and preserve the value of similar commitments it will probably want to make at that time.
Then why not the same of central bankers and their NGDP level target…?
W Peden: “The UK Treasury issued a 51 year bond last year…”
I just can’t see those things possibly finding any buyers! But, of course, I bet they got snapped up. And 51 years is nothing. In the good old days, when the sun never set on the British Empire, they issued consuls. And some of those consuls are still out there, I think, despite government efforts to buy them back!
Dan: you are right, of course. But why couldn’t the same reputational argument be made for monetary policy, as well as fiscal policy? How did the fiscal confidence fairy get reified into a cast-iron bond, while the monetary confidence fairy gets dismissed as just Tinkerbell, who only exists if we all believe in her?
Some in Canada speak of the 2% inflation target as “quasi-constitutional”.
Alex beat me to it!
There is one problem. It’s the problem with all laws, where you want both to give people a commitment about the future, so they can plan, and invest (understood very broadly) today, but at the same time you recognise that sometimes you change your views on what the best law would be, or circumstances change in an unforeseen way. It’s the old problem of rules vs discretion. But most people more or less handle this problem most of the time. You recognise the weight of past commitments, but at the same time recognise that unforeseen stuff happens, and you work out some messy compromise. All you need is a bit of innate conservatism, coupled with a bit of (dare I say “Anglo Saxon”?) pragmatism, and an unwritten constitution works more or less OK.
There are theses things called government bonds, where the government makes a commitment to pay a certain amount to whoever owns them, sometimes 30 years in the future. Some of those bonds are even supposed to be indexed to inflation!!! I can’t see any reason why anybody would believe that commitment. Would it actually be in the government’s interest to do what it said it would do, 30 years previously? That would be most unlikely. After all, it will be a totally different government 30 years from now, and it won’t care what a government 30 years previously had said it wanted it do do. Those government “bonds” must be worthless, right?
Ha! Very nice.
But, it should be noticed that there are lots of governments that are not trusted to honor their commitments in 30 years, and hence cannot issue 30 year bonds. And governments that can issue 30 year bonds, cannot necessarily make other commitments at the same horizon.
Yes, the US constitution says that federal debt must be honored. It also says that confederate debt may not be honored. US bonds are not worthless. Confederate bonds turned out to be. The types of commitments governments can credibly make about future policy are historically and institutionally specific. They can’t be treated in an axiomatic way.
So i disagree with Dan K. here — the fact that governments issue long bonds and honor them cannot be derived from any simple premises. It’s a fact about the world, that could be otherwise. In fact we see multiple equilibria of this kind al the time. A government that could issue 30 year bonds, would not face liquidity constraints and would honor those bonds. but if it can’t, in general, issue 30 year bonds, then if it does manage to sell some, it would be likely to face liquidity constraints before they mature and may well suspend payment. So people are right not to buy them. Both the will-honor and won’t-honr equilibria are stable, and rational agents will choose to behave in a way that maintains whichever one you are in. (We’re seeing a bit of this in Europe right now.)
The fact that (some) government are believed very, very unlikely to default on their bonds is a product of a very long history, of a whole set of institutions gradually developing that reinforce that belief. It’s the very gradualness that makes the commitment credible and the equilibrium stable. Many (most?) governments in the world cannot issue 30 year bonds, let alone consols, certainly not in their own currency. And you can’t change that just by announcing it.
Nick, I think you are right to make an analogy with law and other social norms that have to both be regarded as binding, and be allowed to change over time. But I think you underestimate how hard a problem that is. The fact that after a great deal of social evolution it’s been sort-of solved for some aspects of social life in some parts of the world, does not mean that it has been solved in general
JW: “Many (most?) governments in the world cannot issue 30 year bonds, let alone consols, certainly not in their own currency. And you can’t change that just by announcing it.”
Agreed. But notice the one asymmetry in my analogy: the fear with bonds is always in one direction: that the government will “pay” the debt by inflating the currency. The commitment is always to resist the temptation to inflate. Right now, for the US, for monetary policy, we want a commitment to print more money/create a higher level-path of NGDP/create more seigniorage revenue than people currently expect. “I’m going to be a (slightly) bad boy, I swear!” That’s a lot more credible. (Paul Krugman’s words….damn, I’ve forgotten…..irresponsible.)
“The fact that after a great deal of social evolution it’s been sort-of solved for some aspects of social life in some parts of the world, does not mean that it has been solved in general”
Yep. Agreed. Watch the news. Some societies are really…..not at a good focal point.
Nick, I see what you’re saying, but I have never said that central bank stimulus policy is ineffective – in the places where I think it is ineffective – because the bank’s commitments are not credible. It’s more that I don’t think the actions they are committing to are all that important, and I don’t think the number of people who hear and care about those commitments are all that numerous. If my neighbor says he is going to keep the rate he charges for plowing driveways at $35 for the next 5 years, I might find his commitment 100% credible, but its not going to affect my economic behavior much. Now Ben Bernanke’s most totally believable commitments fall somewhere between the commitments of my neighbor and the covenants of God in their impact. But where exactly they fall is the issue.
I know that the latest wrinkle in thinking – spurred in part by the reaction to Woodford’s paper I guess – is that even though the Fed had already given a fair amount of forward guidance about their expectations of keeping interest rates extremely low, the guidance was long-termed enough, and wasn’t backed up with enough other talk to make people really, really, really believe it. Fair enough. So they have now been even more forward and committed in the guidance. But we’re talking here about interest rates that are already extremely low, and about the difference between an extra year or so. How much more charge does this new statement add to the cattle prod? It seems to me we’re talking about microvolts at this point.
Maybe the added commitment has a marginal impact. Who can say? Or maybe the companies and consumers are actually waiting until they know interest rates have hit bottom, and won’t move until they catch a whiff of a soon-to-come rise in interest rates. In that case, couldn’t the promise of extremely low interest rates further into the future be counterproductive? Beats me. I just wish we could make policy on the basis of something less speculative than these mind-game models where, for every model that points in one direction, there is a similar model pointing in another direction.
I suspect we’ll never really know what the effect of all this Fed communications business really is. The CEO from some giant corporation could take a happy pill tomorrow – completely unrelated to any Fed press conferences or whatnot – and announce that the company is hiring 3000 people to accommodate all the new business it is expecting next year. If the company is prominent enough, the confidence fairy might take wing as others follow the leader, and all the hiring might generate the additional demand needed to self-validate the investment decisions.
Then why not the same of central bankers and their NGDP level target…?
Alex, we seem to be talking at cross-purposes here. It is no part of my line that the Fed’s commitments are not sincere, or that future Fed’s will not be determined live up as best they can to the promises of past Feds. Whether that is the case or not is neither here nor there for me.
I just don’t think the central bank by itself can hit an NGDP target with any degree of accuracy, so I don’t think it matters how sincere and committed the Fed is to hitting such a target, or how concerned future Feds will be to keep the promises of past Feds. To me, its something like the National Weather service announcing a rainfall target. It doesn’t matter how sincere, committed and institutionally stable they are.