« In praise of (a little bit of) fiscal dominance | Main | Today's GDI release: both GDP and the terms of trade are improving »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

And of course, by extension, all interaction between banks and the Govt are completely trivial (except those which remove or undermine the first-loss position of private capital)

Good point, Too Much fed.

I should modify that statement to -

All transactions in which banks operate as principals.

Banks act as agents in clearing customer's deposits with each other. That obviously doesn't create or destroy money.

And banks act as agents for the government in distributing and redeeming currency. The banks have no direct role in creating or destroy money there - they're just passing it through.

Very good point, Too Much Fed. Thank you.

Sorry if this is a repeat.

JKH said: "ALL transactions of ALL type between banks and non-banks create or destroy money at the margin."

Should that be create or destroy currency denominated debt at the margin?

If not, how is currency destroyed?

The commercial banking system is the endogenous system.

The government/central bank is the exogenous system.

At some level, a bank is a bank.

Just that the exogenous system is not revenue constrained.

The endogenous system is - where revenue hits capital constraints (credit revenue; credit equity).

According to MMT, only a sovereign government as currency issuer is able to influence the net financial assets of non-government, which it does through fiscal operations. Government "deficit spending" ( as distinct from non-government expenditure) increases non-government net financial assets, taxation reduces net financial assets.

Government borrowing simply reduces reserves. The CB only controls the overnight interbank rate, e.g., through OMO that manage the cost to banks of borrowing reserves as a required liquidity provision. However, when governments run large deficits, the CB cannot absorb enough of the reserves being generated by deficit spending through OMO alone, since its balance sheet is finite and it cannot issue government debt. Therefore, the CB coordinates with Treasury regarding its needs, and Treasury issues debt as required to remove reserves through bank purchases that exchange interest-free or low interest reserves for interest-bearing Treasuries. This is simply a neutral asset switch, other then for the interest that is injected into the non-government sector, which also increases non-government net financial assets. This is pretty much a daily operation, in which the Fed and Treasury communicate closely. For all practical purposes the Treasury and CB act as a single entity in this regard, since Treasury simply accommodates the CB in its setting the target rate.

Commercial banks control the money supply but cannot increase or decrease non-government. Their activity cannot increase or decrease non-government net financial assets because their operations net to zero.

The sovereign government is the monopoly provider of its currency of issue, which is called high power money (HPM), base money and M0 (zero). Bank money is only involved in M1, M2, and M3. How the non-government system works is the subject of Circuitism. The relation between government and non-government is the subject of Chartalism.

This is the basis of MMT, as I understand it. Hope I got it all correct.

Oops. "Commercial banks control the money supply but cannot increase or decrease non-government" should read, "Commercial banks control the money supply but cannot increase or decrease non-government NFA."

tjfxh: You are correct. JKH's point is that expenditure by the bank (on its own behalf) has an equivalent impact on horizontal money as if it had made a loan, and is subject to the same constraints.

In my mind, I see parallels between this and how the Govt creates (and destroys) NFA equity via spending (and taxation)

tjfxh | December 01, 2009 at 06:30 PM

Looks good to me.

Winterspeak: "In my mind, I see parallels between this and how the Govt creates (and destroys) NFA equity via spending (and taxation)" Herein lies the problem is the horizontal operations of commercial banks are not distinguished from the vertical operations of the government as currency-issuer in relation to non-government as currency-user. The great contribution of MMT lies in providing understanding now the monetary system actually works. This is not theoretical. It is descriptive of fact (financial operations and how they are accounted for).

The term "money" invites confusion in this regard because it has a number of different meanings that are significant to financial operations and national accounting. In modern economies, the currency of issue provides the numeraire. Because the "money" in use is generally a single, universal and exclusive numeraire in modern economies, "money" applies to both government money (M0) and bank money. These are usually not thought of separately other than for professional reasons. But failing to keep the differences straight results in confusion.

Incidentally, "non-government" means domestic or "private," plus foreign operations in the domestic economy, e.g., trade, capital flows, and foreign purchases of government and non-government financial instruments. Foreign transactions generally go through the same clearing procedure as domestic, so it is not usually necessary to distinguish them or treat them separately in general discussions. So it all gets lumped in as "non-government" in contrast to government as a vertical relationship.

An important point that MMT makes is that the government as currency issuer is providing a public service by providing "moneyness." In this sense, government money (currency issuance and management) is a public utility. This is important because many people tend to think that money is generated by the private sector, so that government borrowing competes for loanable funds and crowds out investment, and taxation siphons off money produced by the domestic economy for "redistribution," thereby disadvantaging the "rightful owners." The reality is that deficit spending adds to net financial assets and taxation subtracts from them. It is really only necessary to subtract net financial assets when additions threaten inflation. Otherwise, the government needs to spend in order to balance the public's desire to save, or there will be an output gap and unemployment will rise. It might be argued that S=I, but this only applies "in the long run." The fact is that firms may not to choose to invest all available saving in a timely way, so that these funds do not contribute to aggregate demand. I suspect that this is what someone (Winterspeak?) previously referred to as "dead money." Moreover, a lot of what counts as saving in times like these is being committed to paying down debt as previous consumption.

The basic MMT principle is that government deficits increase non-government surpluses, and vice versa. Therefore, government surpluses lead to the necessity for domestic borrowing if there is a desire to spend, because incomes will not be sufficient to purchase all available goods and services for sale, and this will lead to an output gap, unless the debt load is increased sufficiently to compensate. Increasing debt load is, of course, unsustainable in the long run.

When the government spends into the economy, the Treasury writes checks as currency issuer. It does not depend on prior "saving" nor does it have to borrow or tax to raise the funds. When these checks are cashed in commercial banks, bank reserves increase. This is an increase in M0 (HPM). This how the government creates money as a financial asset of non-government. Physical currency (cash) enters the economy by banks exchanging reserves for Federal Reserve notes and coins minted by the Treasury and provided to the Fed for this purpose, since the Fed doesn't mint coin.

This is really irnonic. Quoting myself, from the 3rd Canadian edition of Mankiw, Kneebone, McKenzie and Rowe (I wrote this bit):

"Central banks can increase the supply of money in circulation by buying something. They can decrease the supply of money by selling something. It really doesn't matter what the Bank of Canada buys or sells. For example, if the Bank of Canada buys a new *computer* for its researchers with $1,000 of newly-printed currency, the firm that sold the computer to the Bank of Canada now holds an extra $1,000 cash, so the money supply increases by $1,000 immediately." (** added)

I also used the computer market operation as the example. JKH would object that the BoC doesn't normally pay cash, which is a legit objection, but it comes to the same thing.

Now JKH is saying that what holds for the BoC, also holds for BMO, which sounds right to me, provided we define "money supply" as M1, not monetary base, of course.

BUT, how does this follow, Winterspeak: "Talking about underlining, in triplicate, the unimportance of reserves!"?

The supply curve of reserves, and hence the price of reserves, will affect whether or not BMO is willing to buy something, whether it be computers or bonds or IOU's.

Winterspeak:

Here's an analogy for the issue of money:

A landowner collects rent of 50 ounces of silver per year. At a market rate of interest of 5%, that makes his land worth 1000 oz. When he buys groceries, he pays with his own IOU, which says "IOU 1 oz." Call them shekels. He accepts his own shekels in payment of rent, so lots of people accept his shekels as money. Since his land is worth 1000 oz., he can issue up to 1000 shekels without fear of them losing value, but after issuing 1000 shekels his net worth will have fallen to zero.

Now he wants to buy more land, worth 2000 oz. He writes up a bond worth 2000 oz and trades it for the land. The land yields rent of 100 oz./ year, which he uses to service the interest on his bond. His net worth is still zero.

Next, he designates a room in his house as his central bank. His central bank prints 2000 more of his shekels and uses them to buy his 2000 oz. bond (which he can now tear up if he wants). His net worth is still zero, and he could, if he wanted, buy back all 3000 of his shekels by selling his 3000 oz. worth of land and canceling his 2000 oz. bond.

Change 'landowner' to 'government' and 'rent' to 'taxes', and you have a pretty good picture of how green paper dollars are issued.

Next, private banks accept 100 of his shekels on deposit, issuing their own IOU's in the form of 100 checking account shekels. Finally, the private banks create 400 additional checking account shekels, which they lend to a customer who offers collateral worth 400 oz in exchange. In spite of all this money-issue by both central and private banks, the shekel is still worth 1 oz. The shekels issued by the landowner are adequately backed by his land, and the shekels issued by the private bank are adequately backed by the private bank's assets.

Note that banks are not all that powerful, that reserves and capital barely matter, and that bank spending is equivalent to bank lending.

tjxfh: "The term "money" invites confusion in this regard because it has a number of different meanings that are significant to financial operations and national accounting. In modern economies, the currency of issue provides the numeraire."

Bu**er the numeraire. "Numeraire" is something economists use, and it's totally arbitrary. We could use "Venus dust" as numeraire; it would make no difference. What you mean is "medium of account" which is the medium in which real people (as opposed to economists and accountants) quote prices. That matters when prices are sticky. But bu**er the medium of account too; what really matters is the medium of exchange. A general glut of goods (such as we have now) is an excess demand for the medium of exchange. To talk of a "general glut" in an economy without a medium of exchange is economic nonsense.

And until you understand the importance of the medium of exchange (and the distinction between a monetary exchange economy vs a barter economy), as well as the distinction between desired and actual S and I, (and the equivalent distinction between economics and accounting) you wont get your following two paragraphs right either.

For example: "The basic MMT principle is that government deficits increase non-government surpluses, and vice versa. Therefore, government surpluses lead to the necessity for domestic borrowing if there is a desire to spend, because incomes will not be sufficient to purchase all available goods and services for sale, and this will lead to an output gap, unless the debt load is increased sufficiently to compensate. Increasing debt load is, of course, unsustainable in the long run."

No, no, no!


Winterspeak:

I know how banks make loans. If you really think that this is a great insight that pokes holes in the monetarist theory, then you just don't understand.

Think about alternative monetary arrangements. 100% reserves for transactions accounts. A frozen quantity of base money. Develop an understanding that covers those possibilities as well as something closer to the status quo.

You asserted that banks must make loans by crediting deposit accounts. Well, that is a very narrow view of what it means to be a bank. That is one way it can happen.

My examples all involved offsetting changes in bank balance sheets that allowed the quantity of monetary liabilities to change in a way different than the quantity of bank loans. None of its was inconsistent with bank loans being created by having funds credited to deposit accounts or having them repaid by debits to deposit accounts.

I teach all of this in introductory macro and then more so in money and banking.

tjxfh: and I would love to take you on with the philosophy of language too ;)

(Off-topic: where where you when I was doing posts on the Principle of Charity, and needed a philosopher of language?)

Each theory creates its own semantic net, depending on the distinctions it thinks are relevant and irrelevant. Sure, I am misusing language from the perspective of accountants' and finance guys' dictionaries. But who gave those guys the authority to write the dictionary?

In this theoretical context, bonds and promissary notes are theoretically equivalent. So are IBM computers, for that matter. They are all just "stuff that banks buy".

(Off-topic again: do *you* understand what I was trying to say in my posts on "churches and central banks", and "interest rate targeting as a social construction"? Because hardly anybody else did.)

Nick, I'm certainly not going to argue with you about the meaning of numeraire. You are the economist, not me. But I'll explain my thinking.

Investopedia: "Numeraire: An economic term that represents a unit of account. In French, the term means "money", "coinage" or "face value".... An example of a numeraire arises when we look at how currencies were valued under the Bretton Woods system during the mid-twentieth century. The U.S. dollar was priced as one-thirty-fifth the price of an ounce of gold. All other currencies were then priced as either a multiple or a fraction of the dollar. In this situation, the U.S. dollar acted as the numeraire because it was fixed to the price of gold."

My thinking is that the old notion of numeraire was in terms of the fixed relationship of the dollar in relation to gold. I assume that now that convertibility is thing of the past, that the dollar itself becomes the numeraire as the reserve currency. I should also say that I've seen it said that in some economies (Scotland) different bank monies acted as multiple numeraires. If I have this wrong, please set me straight.

Nick's post said: "A general glut of goods (such as we have now) is an excess demand for the medium of exchange."

Can it be a shortage of the medium of exchange?

Care to differentiate between currency and currency denominated debt here?

tjfxh: Sorry, I'm getting overly excited this evening. Too many comments coming too fast!

But yes, investopedia isn't quite right. At least, this is not how economists use those words.

When an economist is building a model, he chooses some good to use as numeraire, sets its price equal to 1, and measures all other prices in terms of that good. The economist's choice is arbitrary. It doesn't matter what he chooses (except it may make the math simpler or harder). The good the economist chooses as numeraire need not even exist in his model (hence "Venus dust".

Then there is the good in which people in the real world quote prices. That's the medium of account. The economist often uses the medium of account (if it is defined in his model) as his own numeraire, but only because it leads to less confusion. he could choose anything.

(The unit of account is a particular quantity of the medium of account. If gold is the medium of account, one troy ounce of gold might be the unit of account, for example. Only really prissy economists insist on distinguishing between the medium of account and unit of account.)

But the good which people use to buy or sell all other goods, the medium of exchange, is what is really important. It's what I call "money". It's normally convenient for real people to use the same good as medium of account and medium of exchange, but in some (rare) cases prices are quoted in one good, and payment is made in another.

In Scotland (I used to live there), there was one unit of account (the pound) but multiple media of exchange. But since the Royal Bank of Scotland notes, and Bank of Scotland notes, etc, were all redeemable into bank of England notes, at par, all media of exchange functioned equivalently in Scotland (all were equally acceptable north of the border, but not too far south of the border).

One of the big failures of "MMT" is that it implicitly assumes (quite correctly) a monetary exchange economy (as opposed to a barter economy, or an economy with one centralised auction market), but insists on defining "money" as the medium of account, thereby missing the essential importance of the medium of exchange role of money, even within "MMT". Without a medium of exchange, "MMT" would be theoretically nonsense.

For example, in a barter economy, there cannot be a general excess supply of goods. In a barter economy, an offer to sell one good *is* an offer to buy another good.

But that criticism of MMT would lead me totally off-topic.

Off-topic again: the HG221 shelves of any library are stacked full of dusty tomes with titles including the words "modern" and "money".

Nick, here is a quote from Bill Mitchell that summarizes what I was trying to say, however naively:

The only way the private domestic sector can save if there is a current account deficit is for the government sector to run deficits up to the desired private saving. Government deficits “finance” private saving by ensuring that aggregate spending is sufficient to generate the level of output and income that will bring forth the private desired saving levels.

"Unemployment occurs when net government spending is too low. As a matter of accounting, for aggregate output to be sold, total spending must equal total income (whether actual income generated in production is fully spent or not each period). Involuntary unemployment is idle labour unable to find a buyer at the current money wage. In the absence of government spending, unemployment arises when the private sector, in aggregate, desires to spend less of the monetary unit of account than it earns. Nominal (or real) wage cuts per se do not clear the labour market, unless they somehow eliminate the private sector desire to net save and increase spending. Thus, unemployment occurs when net government spending is too low to accommodate the need to pay taxes and the desire to net save."

"How large should the deficit be? To achieve full employment net government spending has to be equal to the non-government desire to net save to ensure there is no aggregate demand gap."

http://bilbo.economicoutlook.net/blog/?p=5194

tjfxh:

The only way for me to handle Bill's quote is to fisk it:

"Unemployment occurs when net government spending is too low."

Only if you define "too low" in that way. If the price level and rate of interest can and do adjust, you can have a fall in government spending and no change in unemployment. And you can have unemployment that cannot be cured by fiscal policy 9or monetary too).

"As a matter of accounting, for aggregate output to be sold, total spending must equal total income (whether actual income generated in production is fully spent or not each period)."

No. As a matter of accounting, total spending equals total income regardless of whether aggregate output is sold, whether there is mass unemployment, or excess demand for labour.

"Involuntary unemployment is idle labour unable to find a buyer at the current money wage."

OK. Not the same as Keynes' definition, but I can live with it nevertheless.

"In the absence of government spending, unemployment arises when the private sector, in aggregate, desires to spend less of the monetary unit of account than it earns."

No. He has confused medium of exchange with unit of account. And he ignores other possible causes of unemployment (such as efficiency wage theories) that have nothing to do with insufficient aggregate demand.

"Nominal (or real) wage cuts per se do not clear the labour market, unless they somehow eliminate the private sector desire to net save and increase spending."

This assumes that the unemployment is due to deficient aggregate demand. It also ignores investment (unless it defines "net saving" as "saving minus investment").

"Thus, unemployment occurs when net government spending is too low to accommodate the need to pay taxes and the desire to net save."

But this ignores the role of the price level and rate of interest etc in affecting desired saving and investment.

Look. In short, this is basically a mix of accounting identities and the macroeconomics I learned in high school.

Too much Fed:
"Can it be a shortage of the medium of exchange?"

'Excess demand' is economist's speak for 'shortage'.

"Care to differentiate between currency and currency denominated debt here?"

Nope. The key distinction is between 'medium of exchange', which includes currency but also includes chequing accounts; and stuff that isn't medium of exchange.

NIck, thanks for advancing the debate for me. Without clearly understanding all sides, the matter reduces to ideological preference. I'm just in the process of acquiring economic literacy, so have to think about this.

Perhaps if Scott F. is around, he will take up the MMT side in response.

tjfxh: As a first year undergrad I had a slight misadventure and developed a rash over my whole body that kept me from sleeping one night. So I read Lipsey's first year economics text cover to cover. By dawn I was economically literate. I would recommend the same to any intelligent person (just skip the girl, bathtub, and large bottle of herbal shampoo). Any university intro text will do.

"Care to differentiate between currency and currency denominated debt here?"

Nope. The key distinction is between 'medium of exchange', which includes currency but also includes chequing accounts; and stuff that isn't medium of exchange.

So, what about a mortgage when buying a house?

Nick wrote: " And you can have unemployment that cannot be cured by fiscal policy"

Could you give an example?

Too much Fed: a mortgage is not a medium of exchange.

Winslow R.: Efficiency wage models (lots of different models under that heading); minimum wages; monopoly unions; search models; hiring cost models; adverse selection models; probably lots more I've forgotten.

Question for Winterspeak and JKH about whether a change in the price of reserves constitute a supply issue. If I understand correctly, you’ve argued that higher interest rates (FFRs as the benchmark) merely result in higher loan offer rates, and whether this affects the market for bank loans then depends on how these higher rates affect demand from borrowers. Banks remain capable of producing any quantity of loans they desire, subject only to demand at a given price. Thus, reserves are only a demand side issue.
Does your terminology preference here apply outside of the bank loan market? Let’s say there was a terrible wheat harvest, down 50%. But that farmers had an available alternative. At a much higher cost than traditional growing, they can make up for the bad weather and produce enough to match (or exceed) the prior year’s quantity, but at a much higher cost (more labor hours and technology inputs). But the only way quantity sold will match last year’s is if demand was perfectly inelastic or the demand curve shifts, since prices will have to be much higher. Are you okay with calling this change in the pricing curve of wheat a supply issue, or do you also see this as only a demand issue because supply is not physically or legally constrained?

Nick wrote: "Efficiency wage models "

It seems you might consider fiscal policy "counter-productive to an employer's employment goals" rather than deciding it "wouldn't work" to reduce unemployment.

Winslow R.: If you have an efficiency wage model of unemployment, then fiscal policy (or monetary policy) can shift the Aggregate Demand curve, but that will not reduce unemployment. It just creates inflation. It's not that the effect is undesirable to employers. It means there is no effect on unemployment. Zero. And the unemployment in these models is "involuntary" in the normal sense of the word. Some workers are employed, and other identical workers would like to work at that same wage, but are unemployed.

Nick: "If you have an efficiency wage model of unemployment"

After the big "if"

"fiscal policy (or monetary policy) can shift the Aggregate Demand curve,"

yes, but fiscal policy can also directly target the unemployed through an ELR (employer of last resort) another part of MMT avoiding the inflationary consequences of inefficiently shifting the Aggregate Demand curve while still reducing unemployment.

Combination of posts said: "A general glut of goods (such as we have now) is an excess demand for the medium of exchange.

Can it be a shortage of the medium of exchange?

'Excess demand' is economist's speak for 'shortage'.

Care to differentiate between currency and currency denominated debt here?

Nope. The key distinction is between 'medium of exchange', which includes currency but also includes chequing accounts; and stuff that isn't medium of exchange."

I think I'm not asking it right. Let's try it this way. If excess savers don't spend now because they don't need to and other people don't have enough debt free money to spend now and can't afford to borrow more from their credit cards, borrow more against their house, or borrow to spend on a house, what should happen?

I'm thinking about Calculated Risk's post(s) saying housing (mortgage debt and I believe new homes) and retail sales (probably credit card debt) have led the economy out of the past few recessions.

The US Fed is very clear: reserve requirements are irrelevant (basically zero). The mechanism is through clearing balance requirements.

Capital constraints are irrelevant because 1) banks are rarely capital constrained in the short-run and 2) bank capital is endogenous and is supplied by the market as needed.

Banks can have as much capital as they want, but capital cannot meet clearing balance requirements, only reserves can. Therefore, only the monetary base affects the rate-of-interest in the reserves market.

I think the real question you're all dancing around is whether the rate of interest in the reserves market is related to the broader economy.

tjfxh: I quite understand how the banking system works, and the difference between horizontal and vertical money. I was just reacting to the fact that all bank expenditures (when they are acting as principal) increase increase horizontal money just as credit extension does. Wild. I used the term "parallel" for that reason.

NR:"The supply curve of reserves, and hence the price of reserves, will affect whether or not BMO is willing to buy something, whether it be computers or bonds or IOU's."

I knew reserves had no effect on bank's ability to lend, and now I learn they have no impact on any kind of principal spending. And yes they impact whether a bank is willing to buy something, but so do about 1000 other factors.

Bill: "You asserted that banks must make loans by crediting deposit accounts. Well, that is a very narrow view of what it means to be a bank"

Great point! Just last Friday I went down to my local bank to take out a mortgage, and they gave me 300 lbs of freshly slaughtered goat meat in honor of Eid al Adha; ) You are right, my focus on bank loans ending up as credits in a deposit account was far to short sighted!

"2) bank capital is endogenous and is supplied by the market as needed. "

and why did we had some 150 bank failures?


"net save", as I understand it, is equivalent to stuffing cash into a mattress though holding treasury securities also qualifies. Basically storing financial wealth in government created financial assets.

I guess Nick could be right about income = spending if spending on treasury securities, or 'government income', is included which I don't think Bill does.

Kind of complicates things if the government starts selling debt for no other reason than to soak up short-term reserves and increase 'spending' to meet the population's desire to 'net save' :)

Winslow: If you added ELR fiscal policy (money-financed), with a money-wage W, to an efficiency wage model, then this is what would happen.

There would be Kurt E Vonnegut jr "Player Piano" jobs paying W that would give workers exactly the same real wage as unemployment insurance pays today. Nominal wages and prices would rise in the private sector until you had exactly the same excess supply of labour to private sector "real" jobs that you have now. The previously unemployed would be working for the government, but would be equally as miserable as the unemployed are now.

Essentially, it would be like a Dickensian workhouse, just financed by printing money.

That's just the logic of the Shapiro/Stiglitz model.

Winterspeak, I was quite aware that you know all this, and I wasn't writing with you in mind, other than in the first sentence. I was just pointing out that folks often equate similarity with sameness, so that is quite usual to overlook the vertical relationship of government to non-government. Sorry about my lack of clarity of intent if you took the entire comment as directed at you personally.

Winslow:
""net save", as I understand it, is equivalent to stuffing cash into a mattress though holding treasury securities also qualifies. Basically storing financial wealth in government created financial assets."

Yes, I think that's how Bill defines it. Same as "savings minus investment" (for a closed economy).

"I guess Nick could be right about income = spending if spending on treasury securities, or 'government income', is included which I don't think Bill does."

No. I'm adopting the same income=spending that's standard in national income accounting, and that Bill must be adopting too, or else he would be wrong for other reasons.

"Essentially, it would be like a Dickensian workhouse, just financed by printing money."

Wow, that gives me a pretty good understanding of where you stand on the WPA.

Winslow: what's WPA?

And no, it doesn't give you any understanding of where I stand on anything at all.

All you should understand from this exchange is that you are arguing with someone who actually does understand efficiency wage models of unemployment, and how your ELR policy would work if the efficiency wage model were true. Someone, in other words, you has actually made the investment to understand this stuff, and is not just BSing.

Nick wrote: "what's WPA?"
http://en.wikipedia.org/wiki/Works_Progress_Administration

Nick, good night!

If you think I need to review effiency wage models, "if the efficiency wage model were true", please point me in the right direction. I think I understand it pretty well, but I could be wrong.

Morning Winslow! Yes, i should have gone to bed earlier!

At a theoretical level, i really like the ELR policy. Deficient-demand unemployment is caused by a shortage of medium of exchange in the right hands. The ELR automatically puts it in the right hands. It creates a perfectly elastic AD curve, with an immediate and very direct transmission mechanism for what I call monetary policy (and you call fiscal policy, though it's really both). The monetarist helicopter flies over the unemployed. Practical problems, but leave them aside.

But if you match it with an efficiency wage model, you get very weird results. That model implies that there is an equilibrium wage differential over the "outside opportunity". Normally we think of that as unemployment insurance, but with an ELR, it would be replaced by those "Last Resort" jobs.

Now, the government fixes the nominal wage on those last resort jobs, and prints money accordingly. The efficiency wage model determines the real wage, not the nominal wage. That means the price level must rise sufficiently, and the real wage on last resort jobs must fall sufficiently, until the original real wage differential of "good jobs" over last resort jobs returns to what it was originally, as a differential over unemployment insurance. So in the new equilibrium, real wages in last resort jobs must make workers in last resort jobs as miserable as the unemployed were before the ELR program was instituted.

Which is why my "Dickensian workhouse" comment.

With that Shapiro/Stiglitz efficiency wage model, unless there is some policy that can tackle the original cause, there is a sort of "iron law of constant misery" of those not in good jobs. Anything you do to reduce the misery of a worker not in a good job, just causes a proportionate increase in the number not in good jobs. The total misery stays the same.

FYI, from a paper by Randy Wray in 2000 at http://www.cfeps.org/pubs/wp/wp9.html

The first component of the proposal is relatively simple: the government acts as the employer of last resort, offering to hire all the labor that cannot find private sector employment. The government simply announces the wage at which it will hire anyone who wants to work, and then hires all who seek employment at that wage. A package of benefits could include healthcare, childcare, sick leave, vacations, and contributions to Social Security so that years spent in ELR would count toward retirement. Of course, there will still remain many (non-ELR jobs) jobs in the public sector that are not a component of the ELR and that could pay wages above the ELR wage. This policy will as a matter of logic eliminate all unemployment, defined as workers ready, willing and able to work at the going wage but unable to find a job even after looking. Certainly there will still exist many individuals—even those in the labor force—who will be voluntarily unemployed; there will be those who are unwilling to work for the government (perhaps at any wage!—survivalists and the like), those who are unwilling to work for the government's announced wage (for example, because their reservation wage is too high), those who are between jobs and who would prefer to look for a better job while unemployed, and so on.

The ELR will eliminate the need for a minimum wage, as the ELR wage will become an effective minimum wage. It could also establish the base package of benefits that private employers would have to supply. It could replace unemployment compensation, although it could be simply added on to give workers who have lost their jobs more choices. In the US well under half of the officially unemployed even qualify for unemployment compensation. The point is that no matter what social safety net exists, ELR can be added to allow people to choose to work over whatever package of benefits might be made available to those who choose not to work. Obviously, generous benefits to those who do not work can affect willingness to work. The ELR benefit and wage package should be set higher than the benefit package given to similar individuals who do not work, but even this is not absolutely necessary. If ELR enhances one’s access to desirable private and public sector (non-ELR) jobs, then some individuals will choose to work in the ELR program even if this means taking a benefit cut. However, if society values work, it seems far more reasonable to reward ELR workers with a better compensation package than they would receive if they did not work.

NR: "One of the big failures of "MMT" is that it implicitly assumes (quite correctly) a monetary exchange economy (as opposed to a barter economy, or an economy with one centralised auction market), but insists on defining "money" as the medium of account, thereby missing the essential importance of the medium of exchange role of money, even within "MMT". Without a medium of exchange, "MMT" would be theoretically nonsense."

No. MMT is very clear about where it applies and where it does not, and it would not claim to apply to a barter economy. If only monetary theory was as clear and would exclude itself from current reality! There are economies in the world where the Govt is not a currency issuer, and monetarists could make themselves useful limiting their inquiry (and recommendations!) to those.

"and why did we had some 150 bank failures?"

But that misses the point. Just because capital is supplied as needed does not explain which institutions receive the capital. This process depends upon a notion of the return earned on the capital. Once the CBs of the world committed to steady inflation, the risk to capital declined, but certain bank managers show themselves to be a risk... ergo, the capital goes to the other banks (and so does everthing else) and naturally equilibrates to maintain a fair return.

I have a real problem identifying distributed actions as the independent variable. Bank capital allocation is a distributed process driven by the 'price system' and thus must be endogenous in any discussion of the way monetary policy works.

The exogenous variables must be those arising from centralized action because those are the only factors that can plausibly be set arbitrarily.

"With that Shapiro/Stiglitz efficiency wage model, unless there is some policy that can tackle the original cause, there is a sort of "iron law of constant misery" of those not in good jobs. Anything you do to reduce the misery of a worker not in a good job, just causes a proportionate increase in the number not in good jobs. The total misery stays the same."

I don't see how the idea of a "iron law of constant misery" is justified.

As you say, "The ELR automatically puts it[money] in the right hands."

Unless you see that 'misery' comes from earning the lowest wage on the economic ladder or from working for the government, I don't see the allocation or even the quantity of 'misery' remaining unchanged.

The current U.S. system doesn't provide unemployment benefits (or any other benefits for that matter) to the long-term unemployed and are therefore in deep 'misery'.

I'd propose the 'iron law of constant misery', if considered true, would then have a compliment called the 'iron law of constant nirvana'. Taking a bit from those in a state of 'nirvana', using a government transfer, to negate some 'misery' seems to be the crux of the ELR program.

Total misery/nirvana would remain the same, just nirvana wouldn't be as good and misery wouldn't be as bad. Seems a lot like an argument about the efficient allocation of resources without regard to the benefits of a more egalitarian society.

Would you argue that an egalitarian society is inherently inefficient?

If the floor on misery is raised, does the overall societal benefit more than exceed the cost of raising the floor?

Are there 'benefits' to raising the floor?

"Bank capital allocation is a distributed process driven by the 'price system' and thus must be endogenous in any discussion of the way monetary policy works.

The exogenous variables must be those arising from centralized action because those are the only factors that can plausibly be set arbitrarily."

The ratio of bank capital to loans can be set 'exogenously'.

Winslow R: Yes, capital requirements can be set exogenously. TARP in effect relaxed them.

Winterspeak: "No. MMT is very clear about where it applies and where it does not, and it would not claim to apply to a barter economy."

Agreed. But then for consistency they ought to define money as medium of exchange. But, the criticism I am making here of MMT I would also make of many other macroeconomic approaches. MMT is in good (bad) company!

Jon and Winslow: on bank capital. I think you can perhaps make a good argument that the supply curve of bank capital has been vertical (or close to it) recently (during the crisis). That's because bank capital is sort of "locked in". Banks' shareholders can't, in aggregate, pull their capital out of banks. If the price at which the banks could raise new capital would imply a low price at which new shares would be issued, then the supply curve of bank capital would be vertical up to some point, and only then start sloping up. And right now, banks may very well be on the vertical portion of that supply curve. I don't know if that's making sense.

Winslow: "I don't see how the idea of a "iron law of constant misery" is justified."

It's not justified. It is predicted by the model.

The "No shirking condition" in Shapiro/Stiglitz says that the expected benefits from shirking are equal to the expected cost of being caught shirking. That latter is the probability of detection, times the "misery" of being fired if you are caught. So, for given expected benefits, and given probability of detection, that "misery" is a constant. You can think of that "misery" as "misery per week times number of weeks until you get a good job again".

If a nice person does some policy that halves the relative misery per week (give bigger transfers to the unemployed), you just double the number of weeks of misery. If a sadistic person did something to double the misery per week of the unemployed (flogging them), he too would be disappointed to find that the weeks in misery would halve.

Yes, the predictions of that model are indeed miserable.

Bank capital: having another go to explain this clearly:

Since banks can issue new shares, but shareholders can't in aggregate, redeem old shares, the supply curve of bank capital look like a hockey stick, standing upright. A vertical shaft, then an upward sloping blade. In normal times, banks are on the point where the stick curves. A drop in the value of banks' assets shifts the stick horizontally left. A fall in bank share prices shifts the stick vertically up. We have just had both. So that has put banks on the vertical portion of their capital supply curves. That's why JKH's model starts to work in these special circumstances.

MMT has a medium of exchange, it's that which is used to settle a tax liability.

Inquiring minds are asking how, if the existing models that were used recently are so good, what happened. We now have a global financial crisis that has turned into a real crisis of historical proportions. This situation seems to be nowhere near resolved since many economists are increasing the odds for a double dip, with even higher unemployment coming. Even relatively rosy assumptions foresee a new normal with higher unemployment than previously experienced post recovery, coupled with weak growth.

When asked why this wasn't foreseen, the orthodox answer seems to be, "No one could have seen it coming." Needless to say, this is infuriating to a whole lot of people down here in the US, many of whom are suffering, and many of those not suffering yet are getting scared that this things are spinning out of control because the folks running things don't really have a handle on what is happening and are just bailing out their buddies. As a result the political scene here is turning somewhat ugly.

What say?

tjfxh: "Inquiring minds are asking how, if the existing models that were used recently are so good, what happened."

Macro guys ignored finance; finance guys ignored macro; and nobody was watching or understood all the hundreds of things you would need to watch and understand. Sure, that's a massive oversimplification, but if you want my answer in one sentence, that's it.

So, as a macro guy, that's why I'm doing posts like this.

Scott F, I'm glad you posted that Wary quote. I had read that previously and was wondering how practical it is for the to set a floor wage with the type of benefits that Randy mentions, which would really be necessary in the US considering the miserable safety net here, where to qualify for most aid programs one has to be pretty much destitute. I don't get the idea that the solution of people proposing a job guarantee, like Randy and Bill, are thinking of creating a floor at the misery level. I haven't read their technical stuff. Do they present tight arguments overcoming serious economic objections such as Nick raises?

Here's the political and policy issue. Many people, me included, are vitally interested in this debate, but either don't have the economic/mathematical chops to wade into this, or if they do, lack time to give it the attention it requires. That's why a public debate among experts is so important, instead of just pitting one ideology against another, as so often happens. Then, bias, myth, and emotions rule instead of reason and evidence.

To seed this further, let me bring in another criticism, this time from Circuitist Steve Keen instead of Neo-Chartalism. Steve was one of the people who correctly called the comideveloping financial crisis, based on his analysis of debt. According to him and others, like Michael Hudson, the core problem is debt, and it is not yet being addressed in the way it needs to be:

Only one question remains: why do Central Banks ignore the debt to GDP ratio?

There is nothing more dangerous than a bad theory.

The simple reason is: because they are neoclassical economists. You don’t get to be a Central Banker without a degree in economics, and the school of thought that dominates economics today is known as neoclassical economics. Though a lot of what it says appears to be superficially intelligent, almost all of it is intellectual drivel, as I outlined in my book Debunking Economics (which summarised a century of profound critiques of this theory which its practitioners have studiously ignored).

Since critiques by economists and mathematicians of this theory have literally filled books, I won’t try to go into all of them here. Just three key neoclassical myths suffice to explain why they do not understand the dynamics of our credit driven society. They believe that:

(1) The nominal money supply doesn’t affect real economic output;

(2) The private sector is rational while the government sector is not; and

(3) That they can model the economy as if it is in equilibrium.

The first myth means that they ignore money and debt in their mathematical models: most neoclassical models are in “real” terms and completely omit both money and debt. So since debt doesn’t even turn up in their models, they are unaware of its influence (even though their statistical units do a very good job of recording the actual level of debt).

The second myth means that they are quite willing to obsess about government debt, but they implicitly believe that private debt has been incurred for sensible reasons so that it can’t cause any problems.

The third myth means that they ignore evidence that indicates that the economy is very far removed from equilibrium, and they misunderstand the effect of crucial variables in the disequilibrium environment in which we actually live.

Steve Keen
http://www.debtdeflation.com/blogs/2009/12/01/debtwatch-no-41-december-2009-4-years-of-calling-the-gfc/

Nick: Everything I've seen about MMT explicitly describes the world through accounting transactions, and Scott's "settle tax liability" definition is even clearer than that. This also has the nice feature of establishing what drives demand for fiat money, thus preemptively killing the coordination fairy tales so beloved by Austrians. Frankly, I think that is more accurate and easy to understand than a wooly economist definition like "medium of exchange". Maybe economists should describe the world through accounting transactions!

I'd love to refocus, if only briefly, on the original topic of this post. Would you agree with the following statement?

"Bank reserves do not constrain bank lending, but are important in setting the FFR. To the extent that the FFR drives ultimate downstream demand for loans, then reserves matter. If other factors outside of the FFR are playing a bigger role it determining downstream demand for loans, the FFR (and by extension reserves) are less important -- by tautology. Reserves, however, have no impact on the quantity of loans banks can supply, that is driven by capital requirements, although they may influence the cost of that supply. Capital requirements are the operational constraints on the quantity. There is no "multiplier" that takes reserves and automatically ejects them as loans into the economy, which is why a large pool of reserves and shrinking credit are perfectly compatible operational realities"

I think this is the typical MMT position which also allows for the mechanism where reserves do play a role.

Nick wrote: "A fall in bank share prices shifts the stick vertically up. We have just had both. So that has put banks on the vertical portion of their capital supply curves. That's why JKH's model starts to work in these special circumstances."

Are you saying the part of 'JKH's model' that says bank capital will constrain loans, works during these 'special circumstances' when banks have a difficult time raising capital?

If so, what do you believe contrains loans during 'nonspecial circumstances' (when stock can be issued at high prices)?

Perhaps nothing?

tjfxh: the guys running the economy have no idea how a fiat economy works. it's like blaming witch doctors because it didn't rain, (although in this case there's a hose lying right there on the ground which is the real -- pun intended -- tragedy)

tjfxh

good comments, as always. Of course, Michael is an MMT'er, mostly. And we (MMTers) did start predicting all of this actually back in the late 1990s . . see for instance here (many more examples, though): neweconomicperspectives.blogspot.com/2009/08/financial-engineers-and-brave-new-world.html

regarding debate, the problem is mostly ideology, methodology, etc., so it's hard to get it to the level you want. We mostly end up talking past each other or being outright dismissive. Nick's a glowing exception in this regard . . .this is definitely the most I have ever seen a neoclassical engage these alternative views. Even in that case, though, as you've seen, a lot work to do on everyone's part to eventually get a common understanding of what we're talking about in the first place.

Thanks, Scott. As a former philosophy professor and clergy person I'm well aware that simple and even apparently stupid questions are useful in bringing out knowledge. "Out of the mouth of babes." I took Econ 101 (Samuelson's text) over fifty years ago and haven't really looked at this stuff until recently. Mostly owing to the crisis and the policy considerations arising from it, I persist. I'm really happy to have found a place where different sides are discussed rationally and respectfully. The world needs more of this.

Winslow: Capital constraints always limit bank lending, unless the regulators decide not to enforce them.

Banks can always raise money if they can make profitable loans. That is the purpose of banking, to make profitable loans (aka. loans that will be paid back). I think banks may be having trouble raising money right now because:
1) They have all these unrecognized bad assets on their books which, if written down, will eat through their capital and who needs that?
2) Shortage of demand from customers who want to take on debt AND can pay it back

Winterspeak: "Would you [Nick] agree with the following statement?

"Bank reserves do not constrain bank lending, but are important in setting the FFR. To the extent that the FFR drives ultimate downstream demand for loans, then reserves matter. If other factors outside of the FFR are playing a bigger role it determining downstream demand for loans, the FFR (and by extension reserves) are less important -- by tautology. Reserves, however, have no impact on the quantity of loans banks can supply, that is driven by capital requirements, although they may influence the cost of that supply. Capital requirements are the operational constraints on the quantity. There is no "multiplier" that takes reserves and automatically ejects them as loans into the economy, which is why a large pool of reserves and shrinking credit are perfectly compatible operational realities"."

No, not really.

Winslow:
"Are you saying the part of 'JKH's model' that says bank capital will constrain loans, works during these 'special circumstances' when banks have a difficult time raising capital?"
Roughly speaking, probably, yes.

"If so, what do you believe contrains loans during 'nonspecial circumstances' (when stock can be issued at high prices)?"

The reserve supply function of the central bank.

BUT:
1. I would focus more on deposits than loans.
2. Bank loans are only part of the total supply of loans anyway.

""If so, what do you believe constrains loans during 'nonspecial circumstances' (when stock can be issued at high prices)?"

The reserve supply function of the central bank.

BUT:
1. I would focus more on deposits than loans.
2. Bank loans are only part of the total supply of loans anyway. "

I'd say the reserve supply function only works in 'special circumstances' as well.

In particular, when the Fed is willing to invert the yield curve as this starts to pull reserves from the system (FFR acts as a bank tax) perhaps even faster than they are put in (through government interest payments).

Yes, banks can borrow at penalty rates but eventually it starts eating into capital and therefore solvency issues start to arise. Given a bank is leveraged 10:1 a yield inversion can fairly quickly start to reduce bank profitability. Add off-balance sheet entities and the higher leverage ratio/smaller capital base and entities can quickly go insolvent at the periphery.

I sometimes say the only current tool (inverted yield curve) the Fed has to constrain the financial system is to break it. From my perspective the transmission mechanism needs a redesign.

Nick's post said: tjfxh: "Inquiring minds are asking how, if the existing models that were used recently are so good, what happened."

Macro guys ignored finance; finance guys ignored macro; and nobody was watching or understood all the hundreds of things you would need to watch and understand. Sure, that's a massive oversimplification, but if you want my answer in one sentence, that's it.

So, as a macro guy, that's why I'm doing posts like this."

The "neoclassical" models don't really account for currency denominated debt very well. They just assume all the debt was heading for some equilibrium and the fed and the rich are not encouraging it. The people who got it right focused on debt levels, interest payments, wealth/income inequality, and high asset prices.

I can't even get people to admit the difference between price inflating with currency and price inflating with currency denominated debt.

Like, a mortgage is not a medium of exchange, so how does someone buy a house then?

Too much Fed: "Like, a mortgage is not a medium of exchange, so how does someone buy a house then?"
Sorry, but you really need to read a basic economics text. I/we can't be expected to explain everything.

Winslow: normally we say that firms respond to price signals, because the change in price changes the marginal profitability of what they are doing. That's why demand curves slope down and supply curves slope up. We don't normally say that firms only respond to price signals when they are on the brink of bankruptcy. Why should banks be different?

Nick's post said: "Too much Fed: "Like, a mortgage is not a medium of exchange, so how does someone buy a house then?"
Sorry, but you really need to read a basic economics text. I/we can't be expected to explain everything."

Let me try it this way. What % of houses are paid for with all currency?

Thanks for your response Nick. You still believe the "reserve supply function of the central bank" constrained loans during "nonspecial" circumstances."

I guess we're done. Thanks for being a great host!

Thanks Winterspeak! Man, it's hard work arguing with you lot!. But worthwhile, I think. On all sides.

"Only one question remains: why do Central Banks ignore the debt to GDP ratio?

There is nothing more dangerous than a bad theory.

The simple reason is: because they are neoclassical economists."

Why do they ignore debt to almost anything ratio? Because they like the wealth/income inequality (excess savers and excess debtors).

Here is a similar view to Michael Hudson. The fed and the rich want to own all the currency and assets so they can make everyone else rent from them for life (as in no retirement) while the fed and rich can retire at about any time.

I haven't see any neoclassical economists work retirement into their models.

Nick "We don't normally say that firms only respond to price signals when they are on the brink of bankruptcy. Why should banks be different?"

definitely another post....

Winterspeak wrote: [Nick]You still believe the "reserve supply function of the central bank" constrained loans during "nonspecial" circumstances."

Clearly, given the past business cycle, nonbank/bank loans expanded far beyond the desires of a functioning central bank.

Essentially the 'constraining ability' of the reserve supply function didn't take effect until the very end when it was shifted into reverse and everybody quit 'dancing'. I hope Nick appreciates this, though he has also pointed out the capital constraints didn't seem to work so well either, until now as new 'capital' has become more difficult to create.

Winslow: Nick's position to the contrary, there is no constraining ability of the reserve supply function (if by that you mean quantity of reserves. FFR's impact on bank lending via interest rates is, at best confused, at worst inconsequential).

In the past business cycle, I did not see the central bank do anything to try and get a handle on private credit. I currently see them doing their best to reinflate it, so I guess I don't agree that it went "far beyond their desires" since their desire right now is to get right back there!

Capital constraints are only as good as their enforcement. Enforcement before was lax due to explicit support of various off balance sheet vehicles and securitization, enforcement explicitly waived in the days of TARP, and enforcement is a joke in the Obama administration's world of "too big to fail". At least capital requirements work in theory, reserve requirements don't even do that!

Guys: forget bank loans. Loans aren't the only asset on banks' balance sheets. And banks aren't the only source of loans. It's money that matters. Demand deposits are created by banks, and are by far the biggest component of the medium of exchange. That's the only reason that banks really matter. It's got nothing to do with loans, except insofar as loans affect money.

New post coming up soon on this topic. Let's save the argument on this point till then. Keep it all in one new place. Where I'll have you guys attacking me from one side, and all the other economists attacking me from the other side. See if I can survive a "war on two fronts".

Why money and not loans?

Your theory, please.

anon: wait for the post! May take a few days. I have other stuff on.

I’ve been traveling and busy with conferences over the last few days but I just couldn’t prevent myself in getting involved on this issue of money supply behaviour. Your micro model of bank behaviour, whether partial or general equilibrium, remains one in which banks (or loan officers) optimize purely on the basis of the amount of bank assets and reserves. As others have made it very clear, when we are talking macroeconomically of the banking sector as a whole the only constraint is the amount of creditworthy borrowers. Hence, the issue of reserves and capital requirements can only affect the composition of the suppliers of loans but not the overall volume of loans which is determined by demand. Banks as a whole lend as much money as they want as long as there are creditworthy borrowers out there --- that is borrowers who do not jeopardize individual bank solvency.

In any case, I won’t repeat what others have said it better than I could. However, what I was astounded to read was when you wrote on November 29 that: “Post Keynesians literally believe their simple model of the horizontal supply curve of reserves. Orthodox monetarists do not literally believe their simple model of the vertical supply curve of reserves. But PKs seem to think we do literally believe it. That's what's so astounding!” My experience in attending conferences at the CEA, AEA, EEA, etc. over the last 35 years has been that whenever the issue of central bank behaviour has been brought up in the discussion the minority of heterodox economists who used to describe the money supply curve or the supply of reserves as essentially horizontal (like myself) were rejected by the mainstream on both supposed theoretical rigour and on its incompatibility with the real world. If you are correct I must admit that all those debates (including perhaps the present one) have been futile. Do you mean to say that all those guys that convinced the Bank of Canada to go monetarist in 1975 really didn’t believe in their models?! You’ve referred a great deal to the Mankiw, Kneebone, McKenzie and Row book’s analytics of the money supply process, I would like readers who are not satisfied with the mainstream to consider the Baumol, Blinder, Lavoie and Seccareccia book in Macroeconomics in which we have depicted precisely a horizontal supply curve of base money (on page 298) because we do fundamentally believe that it is a more accurate description than that a shifting vertical one!

Hi Mario! Welcome aboard! I was hoping you and Marc might chime in.

Distinguish two questions:

1. The normative one of how the Bank of Canada *ought* to set the money supply and reserve supply curves, from the positive one of how it does set the money supply and reserve supply curves. The 1970's monetarist debate was a normative one. If monetarists had thought the Bank of Canada had in fact been doing what they thought it ought be doing, what were they complaining about?

2. How long is the "run"? Do you really believe the Bank of Canada can and does set a horizontal supply curve of reserves and/or money, and can leave it there at that fixed rate of interest forever? What about the Wicksellian indeterminacy problem? Have you and I not been at PK seminars where models did exactly that, even into the very long run, when stocks of assets adjusted to very long run equilibria (or failed to find an equilibria?

Hey Mario! . . . Yes, your text is the only one that gets it right!

The battle of the textbooks!

If the money supply is determined in a verticalist manner, and held fixed, the AD curve slopes down, and the long run price level is determinate.

If the money supply is determined in a horizontalist manner, and held fixed, the AD curve is vertical, and the long run price level is indeterminate.

Your central bank will either explode or implode, in the long run!

In the long run, a central bank that targets the price level has a long run vertical money supply curve. The Bank of Canada targets the inflation rate. If you ignore the difference between price level and inflation targeting (because it's too tricky for first year), MKMR gets it exactly right. We show exactly what the BoC does (in the long run).

But yes, climbing down a little, it would be nice to distinguish all these different "runs", and do all in the same text. But too complex.

Nick,

Question: Suppose you have a monopoly supplier of some good that it can produce at essentially no cost to itself. Do you doubt that this monopolist can set the nominal price of this good and keep it there for as long as it desires?

Note also that Mario's point about a horizontal supply curve is not suggesting there would be no consequences to setting particular rates of interest, just that at whatever nominal rate is being targeted, the supply curve is horizontal (I said that earlier, actually).

Nick,

1. Your distinction is well taken but not very helpful in this context. The normative question of what the money supply growth ought to be was predicated on the positive one of what they believed was the causal process in place, naturally based on monetarists precepts often supported by questionable empirical work along the lines of Friedman/Schwartz of the previous decade. They accepted the normative position on the basis of an analysis which presumably Gerald Bouey at the time believed to be true! Hence, unless they were misleading him with bad empirical analysis that fit their theories, they must have believed in what they were marketing as theory at the time.
2. The question of the short versus long run is somewhat of a red herring. There has been a long debate going back to the 1980s (especially after Basil Moore’s 1988 book on the so-called horizontalist view) which went on for a while but which Marc Lavoie and I very much believe that it is largely settled. The issue is not that somehow in the “short run” we have the money supply relation horizontal but then it is upward sloping in the “long run”. What happens is simply that in actual calendar time, a central bank may act on the basis of its reaction function if, for instance, it wishes preemptively to combat inflation, say, on the basis of its perception of output growth (as in a standard Taylor rule type model of the reaction function).
3. As to the indeterminacy of the price level, all of that is predicated on the existence of a “natural rate of interest” which both Keynes himself and Post-Keynesians fundamentally reject when coupled with rational expectations which I would hardly subscribe especially in such trying and uncertain times!
4. As to the long-run adjustments of asset values, etc. that is well taken but that only applies to rates of returns on securities, etc. And not on the central bank-determined overnight rate. The latter is not market-determined in the short or longer run, but the spread between, say, the short and long rates can change because of market determined factors.

Scott, I assume the example (in the US) would be when the Fed announces it is changing the FFR. There, there is a "quantum leap," in the rate that the Fed targets and seeks to maintain through OMO.

tjfxh: There's no quantum leap, and sometimes no leap at all. The Fed announces a target, and tries to hit it via OMO. Usually it gets really close. Sometimes it cannot hit it at all. That's one reason why they started paying interest on reserves.

If the Fed used a different mechanism, such as lending unsecured directly to individual banks (which is kind of how the discount window works right now) then it could simply announce the rate and be done with it. Note that there is no interbank lending market in this process. I think that a change in FFR would be a "quantum leap" if implemented via a mechanism like this.

Thanks, winterspeak. That clears up some of the fog in my mind about the "targeting" process. I had assumed that this was a lot more precise than it apparently is in practice, and that the Fed could just jump from one rate to another as it announced its desire.

tjfxh,

You're both right.

The Fed CHANGES the target by simply announcing the new rate, and the rate moves there, as the qty of balances banks want to hold within a given day is very inelastic with respect to the overnight rate. With interest payment at the target, it's even more obvious that they can just announce a new target.

On a day-to-day basis prior to setting the target equal to rate paid on reserves, the Fed used repos (mostly) and reverse repos (less often) to MAINTAIN the existing target, or thereabouts. Now, even this isn't necessary if you keep the system in oversupply and set the target equal to the rate paid. In a crisis like with Lehman where perceived counterparty risk skyrockets, even oversupply may not be sufficient to keep the rate from rising above the target. That's why Warren Mosler and Charles Goodhart both propose setting the remuneration rate at or just below the target, and the cb's lending rate at or just above the target.

Scott:
"Nick,
Question: Suppose you have a monopoly supplier of some good that it can produce at essentially no cost to itself. Do you doubt that this monopolist can set the nominal price of this good and keep it there for as long as it desires?"

No problem, for any regular good. But money is different. First off, what is "the price of money"? If you mean 1/P, where P is the price of goods in terms of money, then yes, the central bank can do this (that's price level targeting). But if by "price of money" you mean the nominal interest rate, then emphatically NO! This is the Wicksell Problem, and it's been known for over a century that central banks cannot do this. (Actually, even David Hume knew this, two centuries ago.)

"Note also that Mario's point about a horizontal supply curve is not suggesting there would be no consequences to setting particular rates of interest, just that at whatever nominal rate is being targeted, the supply curve is horizontal (I said that earlier, actually)."

But it is clear that so many people totally fail to understand what those consequences are. The consequences, in the long run, are a price level of either zero or infinity. I meant what I said about Mario's central bank either imploding or exploding.

Mario: if your model has behavioural functions that are not homogenous of degree zero in nominal variables, it is not well-specified. If it is HD0 then it is, by definition, a natural rate model. And if the central bank does not impose some nominal anchor, by having itself a reserve supply function that is not HD0, then the price level is indeterminate.

Now, everyone reading this might think this is some arcane debate between economic theorists. But I don't believe it is. As I argued in my "social construction" posts, I believe that it is the very fact that people frame monetary policy in terms of the nominal interest rate that is itself responsible for the failure of monetary policy to escape the current recession.

(And yes, I realise I am putting forward a very unorthodox view here, that will raise not just PK eyebrows, but raise the eyebrows of every mainstream economist as well. From my perspective, PKs are just a minor sub-branch of the prevailing Neo-Wicksellian orthodoxy!)

"But if by "price of money" you mean the nominal interest rate, then emphatically NO! This is the Wicksell Problem, and it's been known for over a century that central banks cannot do this. (Actually, even David Hume knew this, two centuries ago.)"

Yes, I mean the nominal, overnight interest rate in the interbank market in which banks borrow/lend reserve balances. If you don't think the cb can set the nominal interest rate that the "money" created only when there are changes to its own balance sheet, then what is the mechanism by which the rate can deviate? In other words, if the Fed sets the rate paid on reserve balances at 0.95% and charges 1.05% for overnight borrowing of reserve balances, and everyone with a reserve account can access these borrowing lending facilities, what is the mechanism by which the nominal, overnight rate for reserve balances would deviate from this range?

And, of course, Hume wasn't working with central banks operating in a flexible exchange rate, fiat money issuing environment.

"If you don't think the cb can set the nominal interest rate that the "money" created only when there are changes to its own balance sheet, then what is the mechanism by which the rate can deviate?"

This is correct, and there is no other answer.

And yes, it is a quantum leap. No qualification required. tjfxh must remain confused...


the Fed started paying interest on reserves for entirely different reasons

the long run is a series of short run targets

"But if by "price of money" you mean the nominal interest rate, then emphatically NO! This is the Wicksell Problem, and it's been known for over a century that central banks cannot do this"

that's exactly what they do

just that periodically, like Keynes, they change their mind

Scott: in the short run, when prices are sticky, central banks can set the nominal interest rate (on short safe assets) to anything they want. But if they hold it there forever, they will either create ever-accelerating inflation, or ever-accelerating deflation. Either way the monetary system collapses. That's the problem.

OK, I'm not going to get into the natural rate stuff for now; note that at 205pm yesterday I said the point isn't that they keep the rate at one place forever, but that they can set it where they want (and anon made the same point).

So, to avoid confusion, I should have said something like . . . if they set a rate on reserve balances at the target rate minus .05% and a lending rate to banks at the target rate plus .05%.

You can assume they adjust the target by whatever optimizing strategy you want. The horizontalist point is that, at whatever rate they are setting, the supply for reserve balances is horizontal. Obviously, that doesn't apply to cases in which the monetary system collapses and nobody holds state's money ceases to be the medium of exchange (well, they could still set the rate, but nobody would hold it). Horizontalists NEVER said the target was horizontal in the long run at only one specific rate; they've ALWAYS said the target rate could be adjusted.

Scott: "Horizontalists NEVER said the target was horizontal in the long run at only one specific rate; they've ALWAYS said the target rate could be adjusted."

Agreed. No sensible horizontalist would say that, and most are sensible. But I think a sensible horizontalist must also say not just that it *could* be adjusted, but that it would *have to* be adjusted in order to prevent inflation from either rising or falling without limit.

Nick,

When I think about it, it is not obvious to me (the discussions on this blog are good for that purpose!) that offering an infinite exchange between money and a certain type of debt at a fixed price (conventionally expressed in terms of an interest rate) does mean that the general price level is indeterminate. For example, if the central bank offers to exchange money for carrots as much as required, isn't the money supply eventually limited by the ability and willingness of the economy to supply carrots? I can see that this would distort the relative price of carrots, but not that it would lead to ever-accelerating inflation.

I look forward to your post on this.

Ok, agreement for the most part on the tactics, at least.

I'll just add that the horizontalist position was always about the tactics, not the strategy of where to set the target and when to change it, so that's why you may not have heard the points about adjusting the rate. While horizontalists like Mario, MMTers, and PK in general, usually don't agree with the natural interest rate concept (there are some exceptions), they do agree that there is a point at which prices would rise and become even indeterminate, and that policy should be adjusted. Some PK think the target rate is what should be adjusted (John Smithin and Tom Palley, for instance), others think that fiscal policy should be adjusted (MMT and some others), some think both or either/or. For MMT, it's the quantity of net financial assets held by the non-govt sector that should be adjusted, since that's "money" for them.

Scott: That is correct. MMTs (like Mosler) would have the rate set at zero, and then use deficits (and surpluses) to control inflation/deflation.

Benefit of this approach is that you take central bank imcompetance out of the picture and focus economic Governance where it should be: leverage in the non-Govt sector (composed of NFA equity, private credit quality, real assets)

"MMTs (like Mosler) would have the rate set at zero, and then use deficits (and surpluses) to control inflation/deflation."

A big advantage of this approach as I understand it is that it breaks the illusion that borrowing and taxing are necessary for deficit spending and kills the "fiscal responsibility meme that dominates American policy. This would allow for a policy that is capable of maintaining full capacity/full employment (figuring 2-3% frictional) along with price stability. This advantage is not lost on Mosler, who is now running for president and needs to relate monetary economics to policy in a way that voters can understand. This means disabusing them of the myths inherent in the established narrative, which are the hold over of a convertible fixed rate system. The MMT position is an upgrade to Abba Lerner's principles of functional finance in light of Nixon closing the gold window on April 15, 1971, upon which the US government became the monopoly provider of a nonconvertible currency of issue within a flexible rate exchange monetary system. See "Functional Finance and Full Employment: Lessons from Lerner for Today?" by Mathew Forstater.
http://www.levy.org/pubs/wp272.pdf

Since government debt issuance is actually for the purpose of draining excess reserves beyond the ability of OMO to handle, in order to allow the Fed to manage its target rate, borrowing would not be needed to drain reserves, and the speciuos link between borrowing and spending could be broken in the popular mindset. Rather, government would run deficits to offset the public's desire to net save, thereby reducing AG below the level necessary to support full capacity, resulting in an output gap and unemployment. Taxation would only be necessary to reduce non-government net financial assets as necessary to prevent inflation from taking hold as the public's desire to save diminished and spending cuts alone could not accommodate this shift in liquidity preference in relation to consumption.

Rebel: if the central banks buys and sells unlimited carrots at a fixed price, the price level is determinate. And it can keep that price fixed forever. Same as the gold standard, only carrots. Same as labour as well (as in some MMTers). All possible (though may or may not be desirable).

But interest rates are different. The central bank can fix something that has $ in the units. Interest rates (nominal and real) have the units 1/time.

My (eventual) post won't be on this topic. Probably on "Why bad banks are a problem".

Winterspeak . . correct, of course. Our Minskyan roots lead us to believe that manipulating interest rates is destabilizing at precisely the moment that you want to generate a soft landing.

tjfxh . . . that's a great paper by Mat. I've used it in class before.

RE: Did you get that? Something the central banks have limited supplies of (gold, carrots) they can buy and sell in unlimited quantities to keep a price fixed.

But they can't punch a number into an excel spreadsheet. Because they can't punch a different number into the same spreadsheet later on. Or.... something.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad