I think we are witnessing the biggest silent shift in macroeconomic thought since the Second World War. For 70 years we have taught, and believed, that we would never again need to suffer a persistent shortage of demand. We promised ourselves the 1930's were behind us. We knew how to increase demand, and would do it if we needed to.
The orthodox have lost faith in that promise; only the heterodox still believe it. And the heterodox have nothing in common, except for keeping the faith.
The orthodox haven't lost hope. They hope that monetary and fiscal policy will be enough to get us out of this recession, and that the limits on monetary and fiscal policy will not be binding this time around. And they are probably right. But they have lost faith that monetary and/or fiscal policy will always be enough – that there are no limits.
And if the Eurozone too turns Japanese, they may start to lose even that hope.
There are two types of macroeconomist.
The first says "What do you mean you can't increase aggregate demand? You run out of paper? Ink? You scared of inflation?"
The second says "But monetary policy won't work at the zero lower bound. And there are limits on fiscal policy, because we daren't let the national debt get too big."
Scott Sumner and Modern Monetary Theorists are examples of the first type of macroeconomist. They have nothing in common, except that one thing. But that one thing is more important than all their differences. And they are heterodox.
Traditional Keynesians and monetarists, the competing schools of the old orthodoxy, belonged to the first type of macroeconomist. They differed only on tactics. They kept the faith. But they have now gone, and only monetary cranks sing the old religion.
The second type of macroeconomist represents the new orthodoxy. Few orthodox macroeconomists today will admit point-blank to having lost the faith, any more than a bishop, no matter how liberal, will admit to being an atheist. But if you believe that monetary policy is ineffective at the zero bound, and that there are limits to how long you can have a big fiscal deficit, it comes to the same thing. You have lost faith that you can always and everywhere increase demand by whatever it takes for as long as is needed.
Losing faith in monetary and fiscal policy, the orthodox turn to financial policy. "If we had better regulation and/or supervision of financial markets and institutions, we wouldn't have gotten into this mess in the first place". That's probably true, but it's also a distraction from the loss of faith. Financial markets and institutions are inherently unstable. They borrow short and lend long; they borrow safe and lend risky; they borrow liquid and lend illiquid; they borrow simple and they lend complex. Finance is magic; you know it can't really be done. Regulation and supervision can never eliminate financial instability. If your faith is contingent on being able to prevent financial crises, you have lost the faith.
Good financial regulation and supervision are important in their own right. A good financial system will better serve the interests of borrowers and lenders. It will create benefits on the supply side. And financial crises will almost certainly cause demand to fall. But just because something causes demand to fall doesn't mean monetary and fiscal policy can't work. The whole point of Keynesian policy was that when (not if) something did cause demand to fall, monetary or fiscal policy could and should be used to increase it back again.
Even suppose the financial system totally collapsed. Why should that prevent monetary and fiscal policy working to increase demand? The biggest flaw of orthodox macroeconomic models is that they have no financial sector. So, if the financial system disappeared, that ought to mean those models would work even better.
This is what I sensed to be the overarching but unspoken theme of a conference at Carleton on the economic and financial crisis. I may do subsequent posts on more specific topics.
Worthwhile Canadian Initiative 
“Actually, this is much the same for inflation targeting”
I think that’s the point I’m interested in
is it the case then that central banks really can’t implement a gold standard without using interest rates as a tool?
and don’t interest rates then become equally important in managing a gold standard as they are without it?
You can still have a gold standard in a world with no borrowing or lending, and so no interest rates. Just use gold coins, for example.
You can imagine a central bank targeting inflation without using interest rates. Central banks control the supply of base money by buying and selling things. Most (nearly all) the things they buy and sell are interest-bearing IOUs (aka bonds, or loans). But they could buy and sell stocks, land, houses, bricks, oil, wheat, anything really (though some things are a lot more practical to buy and sell than others).
“You can imagine a central bank targeting inflation without using interest rates. Central banks control the supply of base money by buying and selling things. Most (nearly all) the things they buy and sell are interest-bearing IOUs (aka bonds, or loans). But they could buy and sell stocks, land, houses, bricks, oil, wheat, anything really (though some things are a lot more practical to buy and sell than others).”
Hmm, I think this is a big difference. When the government buys goods, then it is increasing people’s incomes. When the government buys bonds, it is not increasing people’s incomes, but changing the portfolio composition of their financial assets. Unless there was some real need for the private sector to change it’s portfolio — which seems strange, as these assets trade at indifference prices — then why would this alter household behavior?
On the other hand, businesses receiving large purchase orders from the government will change their behavior — they will hire more workers, invest more (assuming they believe the purchase orders are not one time blips) and create more financial assets (sell more bonds and stocks). With excess capacity, increasing deficit spending on goods will result in an increase in output. But swapping one financial asset for another at fair market prices — at indifference prices — will not result in an increase in output. Why do you believe that it will?
RSJ: If the good that the central bank bought was traded in a competitive market, at a flexible price, and the central bank bought the good at that market price (and didn’t buy so much of it to influence the price), it should be the same as buying a bond. Because the price of the good will equal marginal cost.
If the price of the good were (for whatever reason) above the competitive equilibrium, that will matter.
But I think you are missing the main point that gives central banks their power. They don’t just swap one good/asset for another good/asset. They swap one good/asset for money. Money is special. It’s different. It’s the medium of exchange. All other goods/assets trade against money, and only money. And it is only because we live in a monetary exchange economy, where money is special, that general gluts are possible.
Nick,
“If the good that the central bank bought was traded in a competitive market, at a flexible price, and the central bank bought the good at that market price (and didn’t buy so much of it to influence the price), it should be the same as buying a bond. Because the price of the good will equal marginal cost.”
No, because you still have income effects. If there is unemployment, then you will increase output (and consumption) by purchasing more of the good. Of course, you can argue that in perfect markets, there is no unemployment, so at the end of the day, the belief that deficit spending increases output relies on downward sloping demand curves and excess capacity.
But what is the downward sloping demand curve for bonds? How many more people will be hired to make the bonds, in parallel with the number of people hired to make the bread that you buy? The parallel argument doesn’t hold, and you need to come up with a different channel, rather than arguing that financial assets are goods with a downward sloping demand curve, the manufacture of which increases employment and output.
“They swap one good/asset for money. Money is special. It’s different.”
Nick, you should be able to articulate a channel whereby an increase in government purchases of bonds results in an increase in employment and output. Merely declaring money to be “special” is not enough.
It’s clear that there is a channel by which lowering short term rates can boost employment and output — but this channel is fragile, and is not symmetric. I.e. artificially hiking short term rates causes more damage to output than artificially lowering short term rates. Moreover purchasing more long term bonds does not have the same effect on output as purchasing more short term bonds. And the channel is not symmetric in it’s effects on different sectors of the economy — businesses do not borrow more when short term rates fall, but households can borrow more on real estate. But you cannot keep using that channel — the more you use it, the less effective it becomes.
So this channel is complex and needs to be understood. In order to do that, you do need to get into some of the operational details of the financial system, as well as understand the institutional frameworks.
For example, you need to know that businesses obtain inventory financing at one rate, but long term capital investments are financed at another rate. They tend to be demand-inelastic for the rates charged for working capital, but not long term capital. And there are rapidly diminishing returns to inducing businesses to make long term capital investments as a result of cheaper working capital financing costs.
But you wont see any of that if you pretend that that there is only a single “interest rate”, that financial assets are the same as goods, and that there is a downward sloping demand curve for both. You have to get into the hairy details or else you wont be able to formulate a model that explains why Japan has been stuck in ZIRP for two decades.
Repeatedly making appeals to the “specialness” of money doesn’t add any explanatory power to why monetary policy works sometimes, but not at other times; and it doesn’t explain to what degree monetary policy works — i.e. is it better to induce households to borrow more, or is it better to get the government to deficit spend. It doesn’t explain why cutting short term rates when households have little debt stimulates the economy more than cutting short term rates when households have excess debt.
RSJ: “Merely declaring money to be “special” is not enough.”
I did a number of posts a while back on why money is special. I’m not just saying it is. I just don’t have the metal energy to go through it all again. Search on “Say’s Law” and “excess demand for money” should find you a few, IIRC.
The arguments seem very theoretical, when the issues are surely more contingent. Australia was able to push cash to consumers quickly because it has a relatively efficient central administration with lots of practice in administering handouts, a disciplined political process and had no government debt. Surely none of these conditions hold for the US (or Greece).
More largely, the central issue is a large volume of promises (debt) which cannot be met from the physical economy. At bottom, the central question is political – who is to pay? This is not just a question of power, but it’s surely not just a question of theoretical flows in an ideal world.
And one can agree that money is special while observing that the real economy (of things made, used, consumed etc) is essentially not monetary but physical. It’s the varying linkages between the fiction of money and the physical world that produce the kinds of problems discussed.
I know the argument. As things like income inequality is the refuge of those who have nothing better to say, let’s posit a huge, irrational, sudden demand to stockpile cash under the mattress.
In that case, the central bank will buy bonds, replacing them with cash. The holders of the liquid, riskless, interest bearing government bonds — who all this time wanted to instead hold cash, but forgot that they were able to sell the bonds — now gratefully sell their bonds to the government, receiving cash. They then put the cash under the mattress.
You can tell that they wanted to dispose of their government bond holdings by observing the falling yields.
Once enough of this occurs and the mattresses are sufficiently stuffed with cash — the “desire to hold money balances” is satiated, and demand increases.
All that was needed for this to happen was for the government to engineer this portfolio shift. No cash needed to be transferred from those with differing propensities to consume. No worker needed an increase in their wages or purchasing power, and no business needed an increase in their sales. That all happened endogenously as soon as the central bank bought treasuries for cash, as cash is special.
The problem is that when you try to point out the absurdity of this argument, you are accused of obfuscating the theoretical purity of the model with the thickets of operational details. There is a channel by which monetary policy can work, but the specialness of money is not at play in that channel. It may have worked like this under other institutional contexts — to a limited degree — for example, prior to deposit insurance and central bank liquidity provisioning — but this channel is not relevant to describing the situation of Japan today, or what the U.S. is facing.
And in that case, there is a big difference between replacing a government bond (whose price is rising) with cash, and replacing goods (whose prices are falling) with cash.
“… articulate a channel whereby an increase in government purchases of bonds results in an increase in employment and output. Merely declaring money to be “special” is not enough.”
RSJ, is it really that hard to imagine? Almost by definition, buyers of bonds move money from people/firms who are unlikely to do anything with it to people/firms who are going to spend/invest. If buyers of bonds decide to keep their cash rather than buy a bond, or decide to start demanding their cash back, then spending/consumption and investment will necessarily decline for want of available cash. In extreme cases (e.g. fall 2008/winter 2009), the resulting liquidity crisis can quickly become a solvency crisis (how many businesses where forced to shutdown not because they weren’t viable, but rather because nobody would lend them money?)
If the government steps into the breach and buys the bonds/CP/whatever, then the flow of money from those who would otherwise sit on it to those who are going to spend it/invest it/pay wages etc continues.
Didn’t we cover this in the Bagehot reading assignment? 😉
Nick: Lets start with some simple ideas.
1) In a long-run sense, money is neutral. It must be neutral because its just ‘units’. Changing units, e.g., grams to kg etc cannot sensibly matter.
2) The effect of policy is shaped by expectations. This allows an otherwise neutral nominal change to induce a change in real variables. For instance, for many years of the past decade, real-rates have been negative in the US. That’s only possible because despite setting very low nominal-rates relative to the inflation-rate–and declaring that those low nominal-rates would persist for an extended period of time, inflation expectations did not change.
Expectations seem more stubborn that prices which are contractually sticky more than socially sticky! Consider, for instance, the euro. Its been on a slow, gradual slide down. The major investment banks are estimating a bottom of 1.10:$1. Why doesn’t the the price move more quickly? Expectations are slow to break.
Without expectations, you’d expect accelerating inflation to impinge upon the CB to raise rates to the natural level. Yet, that’s not what happened for most of the decade.
3) Some observers have expressed confusion about some recent remarks from Bernanke, in particular when he was asked by Brad DeLong, “Why haven’t you adopted a 3% per year inflation target?” and answered
That his line of comment seemed so mysterious to the money-is-neutral, its just nominal growth-rates crowd is a strong tell of how embedded expectations are in the psyche.
4) If you look a plot of recent real-rates you’ll see that the Fed managed to work them negative again around mid-2008. Now at this point they almost lost control of inflation expectations. Commodity prices rose rapidly. All of which you can see here: http://lostdollars.org/static/realrate.png The Fed started to cut-rates and drove the real-rate down, but notice what starts to happen, the Fed started to try to decelerate but real-rates continued to plunge because inflation expectations began to break-down. i.e., we were at the cusp of a hyper-inflationary collapse.
But a lot of bad news followed. The bank system spurted and the Fed got a ‘gift’. Inflation expectations broke, real-rates surged. Then panic, and real-rates surged positive again.
At this point in the story, certain growth-targeting economists start wondering why the Fed didn’t pour in the medicine, but I think that misapprehends what the CB saw, namely that inflation expectations had nearly just unraveled.
They got the helm, convinced the world of deflation (again, although it never happened) and now the recipe is playing out (again).
5) These negative real-rates have side-effects, which is to be expected, but one side-effect is questionable: http://www.federalreserve.gov/releases/g17/current/g17.pdf
These periods of negative-rates lead periods of depressed growth in the real capital stock. e.g., as shown on page 5.
This leads to the second problem with a higher-inflation target: real-rates are more likely to be depressed below the natural-rate, ceteris paribus, and its simply not possible to move from a low nominal-rate regime to a high nominal-rate regime necessary to sterilize this effect without blowing a hole in the banking sector. (The relative yields on assets and liabilities would reverse position in the short-run). The alternative escape would be a 70s style inflation. … which could only be undone with a strong recession–again because there would be a loss of confidence in the CB to maintain the inflation-target.
No matter which way you cut it, a recession will come. The only way to avoid it is to settle into a position and maneuver down a narrow channel. History has proved this to be difficult.
A test
Nick Rowe,
You are correct to talk about “faith”. Are the constraints facing policy caused by involuntary system requirements? Or voluntary political and ideological impression imposed, based on “rational” expectations derived by mainstream theory? The last promoted by politicians, rating companies, mainstream economists and speculators that, before economic relations come into play, they impose constraints that become selffullfiling with feedback loops!Is the faith you are talking about based on voluntary requirements?
RJS,
In the presence of pure uncertainty, you protect against ignorance of the situation you are facing and the hesitation of what to do by holding liquidity.This can be in any instrument that requires no spending and employment commitment of resources, such as reserves, currency and highly liquid government bonds as you seem to favor.This point is nothing new or irrational curiosity but proposed by Keynes, Kaldor, the Radcliffe Report, Davidson, Minsky, V. Chick, Moore and many other PostKeynesians. Even Circuit theorists in their recent synthesis accept this point, as a flow constraint not leading to credit, spending and employment.
Patrick,
Yes you are talking about central bank operations to stop a financial panic. There are sellers with no bids, etc, bank runs, failures in the payment system. Lombard Street to the rescue!
But that problem is solved — can be solved with reasonably bright central bankers. The economic and balance sheet problems of the non-financial sector remain. So once you stop the panic, will additional OMO or emergency lending lower unemployment and increase output?
Come to think of it, I think some people are implicitly assuming that deflationary episodes are financial panics.
” So once you stop the panic, will additional OMO or emergency lending lower unemployment and increase output?”
Geez, I dunno. FWIW, I’m with those who say that a direct gift to households of newly printed would to the trick, though it seems that CBs are not inclined to do this.
“eez, I dunno. FWIW, I’m with those who say that a direct gift to households of newly printed would to the trick, though it seems that CBs are not inclined to do this.”
That would definitely do the trick, but it’s fiscal policy, not monetary policy.
An add on to my previous comment. Recent empirical evidence shows a relationship between uncertainty and hoarding liquidity in the form of excess reserves and cash.A simple observation of bank behavior that reduce credit commitments and hoard excess reserves during the recent financial crisis will also show that. Similarly, during periods of uncertainty there is a liquidation of investment positions, leading to rising rates. Furthermore, there is a “pecking order” of liquidity risk and during uncertainty shocks there is a drive up the ladder seeking to avoid the rising flood of ignorance. Only time can reduce/decay this effect. I have attempted to specify this uncertainty effect with a Bernoulli equation that can be reduced to a corresponding linear differential equation whose general solution of the shift factor is a function of time.
Nick Rowe,
Regarding the issue of “faith” in the theoretical analysis of policy. Economists usually estimate technoeconomic parameters upon which they base their forecasts of any feedback effects. However, they fail to comprehend that market participants are also civic units that participate in communities that use ethical criteria of behavior. They can consider policy measures as unfair decisions and unequitable practices and they can react emotionally with demonstrations, riots, strikes, etc. This behavior is correlated with market private behavior and bring a feedback effect that amplifies negatively the technoeconomic parameters constraining spending and income and can bring spiral loops. Argentina and Greece currently are examples altough there many more historically. The crisis that results, does not always leads to stagnation but many times in revolt and a regime switch. These relations were examined by classical economists that correctly thought of themselves as political economists something forgotten by modern economists
Panayotis: “These relations were examined by classical economists that correctly thought of themselves as political economists something forgotten by modern economists.”
The “new institutionalist” economists didn’t forget it. I didn’t forget it. I tried to do my PhD thesis on something very similar. “What happens when the rules of the game, which are themselves constituted by people’s actions and expectations, are endogenous to what happens in the game’s equilibrium?” (I can’t say I got very far answering it.)
I agree about institutionalist economists although they do not go far enough. My point of departure is to introduce another form of analysis for public relations of civic units operating in a community setting induced by the incentive of duty subject to solidarity and cooperation. The criterion of behavior for these units is ethics or decision fairness and equity of praxis. The reaction of these units is expressed with emotion whode feedback amplifies the private unit behavior in markets whose criterion is reason (decision rationality, praxis rule)induced by the incentive of interest subject to competition and synergy. Obviously, an economic unit is the joint manifestation of a civic (public) and shelf (private) side whose intensity of orientation depends on a anumber of factors that I cannot summarize here. The feedback dynamics of a policy such as fiscal consolidation can become closer to the occurrence, while the forecast reality of economic analysis can be seriously surprised!
Isn’t that the point of the Old Orthodoxy, to say that we would never live through another Depression and by extension another World War?
When demand for money goes up (as store-of-value-money-substitutes crash) Hayekians have an argument justifying expanding the supply .. what would be the Rowian suggestion?
“Greg: I tend to associate Austrians only with the view that monetary policy was too loose, rather than adding the view that it is currently too tight, and can and should be loosened. But I am happy to add you to the honourable band of “monetary cranks” (don’t take that the wrong way) if you think that’s an accurate reflection of Austrian views.”
Nick Rowe is off canoing until Mon/Tues. He asked me to let people know why he’s not responding to comments.
Not always and everywhere. Not following thirty years of non-stop keynesian stimulus (except for a brief pause under Clinton), under the banner of Reaganomics.
Eventually it’s gonna lose its oomph.
And it did. Big time. Just in time to hand the whole effed-up thing off to the Democrats to fix again…
With all the failures of monetary and fiscal policy, perhaps we can expect the “mainstream” to grow up and learn that their models aren’t worth two cents. Maybe they will then argue simply for regulations, and then find out that they cannot control anything with regulations, and perhaps embrace a Rothbardian critique of government. The Austrians returning into “mainstream” is coming.
Nick: Enjoy the canoeing!