Keynes' GT Chapter 3

Keynes' "aggregate supply function" in chapter 3 of the General Theory is just the "classical" labour demand function plus the "classical" production function. Except for the weird presentation, there is nothing new there. It is old and boring. It is Keynes' "aggregate demand function" that is new and exciting.

[Update: See Roger Farmer's response.]

Start with three equations:

1. The production function: Y=F(L). Output (Y) is a function of employment (L).

2. A "classical" labour demand curve: W/P=MPL(L). The real wage (W/P) equals the Marginal Product of Labour, which is a decreasing function of employment. This is Keynes' "first classical postulate", which he agreed with.

3. A "classical" labour supply curve: W/P=MRS(L,Y). The real wage equals the Marginal Rate of Substitution between labour (or leisure) and output (or consumption). This is Keynes' "second classical postulate", which he disagreed with (except at "full employment").

From 1 and 2, plus some tedious math, we can derive what Keynes calls "the aggregate supply function": PY/W = S(L). It shows the value of output, measured in wage units, as a function of employment. It is substantively identical to the Short Run Aggregate Supply Curve in intermediate macro textbooks that assume sticky nominal wages: Y=H(P/W), which uses the exact same equations 1 and 2, but presents the same solution differently.

From 1 and 3, plus some tedious math, we can derive a second "aggregate supply function", that is not in the General Theory: PY/W = Z(L). It is substantively identical to the short run aggregate supply curve implicit in New Keynesian models, which assume sticky P and perfectly flexible W, so the economy is always on the labour supply curve and always on the production function.

From 1 and 2 and 3, plus some tedious math, we can solve for Y, L, and W/P, and derive a third aggregate supply function: Y=Y*. This is the textbook Long Run Aggregate Supply curve. It is identical to the solution we could get if we solved for the levels of Y, L, and W/P that satisfied both the first and second "aggregate supply functions".

For example:

1. Y=log(L) production function

2. W/P=1/L "classical" labour demand function compatible with the above production function

3. W/P=Y/(1-L) "classical" labour supply function assuming utility is Cobb-Douglas in leisure and consumption.

From 1 and 2 we get YP/W=L.log(L) which is Keynes' "aggregate supply function". But this is substantively identical to the textbook SRAS function Y=log(P/W).

From 1 and 3 we get W/P=Y/(1-antilog(Y)), which tells us what is happening to the real wage in the New Keynesian model. And we could (someone could) rearrange it to solve for Y as a function of W/P, which is the second "aggregate supply function".

From 1, 2, and 3 together we get Y=(1-antilog(Y))/antilog(Y), which can be solved for Y to get the LRAS curve.

The first aggregate supply curve tells us the output that would be produced if firms could sell the output they wanted to sell, and could buy the labour they wanted to buy, as a function of W/P.

The second aggregate supply curve tells us the output that would be produced if households could sell the labour they wanted to sell, and could buy the output they wanted to buy, as a function of W/P.

The third aggregate supply curve tells us the output that would be produced if both firms and households could sell what they wanted to sell, where W/P adjusts so that both firms and households agree on how much they want to sell.

There is absolutely nothing new on the supply-side in chapter 3 of the General Theory.

It is the demand-side that is new. It is the idea that the demand for goods is a function of the quantity of labour that households are actually able to sell. If households are rationed in the labour market, that will spillover and affect their demand in the output market. Because the amount of labour they are actually able to sell, and hence the income they will earn from wages plus non-wages, depends on demand. Which means that demand depends upon demand. Demand depends on itself. That was new, and interesting.

[Update: but Keyne's weird habit of measuring output in wage units had an unfortunate result: because if YP/W=D(L) (which is Keyne's "aggregate demand function"), we can re-write that as Y=(W/P)D(L), and as Y=(W/P)D(F^-1(Y)), which implies that doubling the real wage, for a given level of output and real income, would exactly double output demanded. Which contradicts Keynes' own consumption function, and only makes sense in a world where capitalists never spend any of their income, on either consumption of investment. Which makes no sense.

I can't help but think that if Keynes had assumed a simple haircut economy composed of self-employed hairdressers, where the production function is Y=L, the whole thing would have been a lot simpler and clearer. Because then W and P would be the same thing, and workers and firms would be the same thing, and the output supply function and the labour supply function would be the same thing, and my three aggregate supply functions would be the same thing. Then he could have concentrated on the output demand function, where he had something new to say, and it would be obvious that unemployment would not be caused by real wages being too high.]

Comments

[Kevin Donoghue]

My only quibble is that "demand depends on itself" is liable to mislead. It sounds like saying: given the value of x, the value of x is determined. In other words, f(x):= x is a perfectly respectable definition of a function. Alas, that's just the sort of remark which (although it's trivially true) leads people into the trap of "Say's Law" (the crude version which I'd like to see renamed the Cochrane-Fama Fallacy but I guess it's too late for that).

[M.R.]

So, on the demand side, the Keynesian idea goes something like this?:

-- output is constrained by demand for goods and services;
-- demand for goods and services is constrained by the level of employment; and
-- the level of employment is constrained by output.

I think that's another way of saying "demand depends on itself."

Under this view, even if prices are perfectly flexible, the economy has no automatic tendency to self-equilibrate back to full employment, unless you believe in a very strong Pigou effect (real balance effect).

[Nick Rowe]

Kevin: fair enough. We have to distinguish the individual from the aggregate thought experiment, and talk about the 45 degree line, to make it clear.

M.R.: I would put it like this: output sold is determined by demand for output, if there is excess supply of output. Demand for output depends on output sold.

Whether or not price flexibility creates an automatic tendency to self-equilibrate to full-employment depends on how changes in prices affect demand, which depends on monetary policy, and anything that could be achieved by flexible prices could equally well be achieved by a sensible monetary policy, only quicker and better. Because there are two ways to increase M/P.

[JW Mason]

Keynes said later that accepting the first classical postulate was a mistake. See for instance his 1939 Economic Journal article, where he accepts Viner's criticism on this point and says that this "is the portion of my book which most needs to be revised."

(Also: How do you write so much? Are pharmaceuticals involved?)

[Edward Lambert]

"It is the idea that the demand for goods is a function of the quantity of labour that households are actually able to sell. If households are rationed in the labour market..."
What would you say to measuring this quantity with the % labour share of income/output?

[Nick Rowe]

JW: I don't remember his 39 EJ. But yes that makes sense. For a given level of demand, and hence a given level of Y and L on the production function, it really doesn't matter what W/P is, provided it is not greater than MPL and not less than MRS. His "aggregate supply function" simply doesn't matter (much). If demand is too low, there will be either excess supply of output (W/P < MPL) or excess supply of labour (W/P > MRS) or, more likely, both at the same time.

I spend a lot of my time thinking about and writing posts. Remember you're a young guy, who has lots of other things you need to do. I'm stunned at how people like Mark Thoma, Scott Sumner, and Noah Smith, can write so much.

[Min]

"If households are rationed in the labour market, that will spillover and affect their demand in the output market. Because the amount of labour they are actually able to sell, and hence the income they will earn from wages plus non-wages, depends on demand. Which means that demand depends upon demand. Demand depends on itself. That was new, and interesting."

Verrrry interesting. :)

But inaccurate. There is a time lag, so that demand(t1) depends upon demand(t0), where t1 is later than t0. Loosely speaking, we may say that demand depends upon itself, but if we are going to reason correctly about demand, we need to be precise. There is a positive feedback loop, but it is not instantaneous.

[Nick Rowe]

Edward: I would say that only makes sense under extreme Kalecksian assumptions that all labour income is spent and none saved, and all non-labour income is saved and non spent. And I think that assumption is very very false. If my salary were $100 lower, and my rent cheque were $100 higher, it would affect my consumption or investment decisions very little.

[Nick Rowe]

Min: it might lag. It might even lead. It depends on expectations.

"Verrrry interesting."

You aren't old enough to remember Rowan and Martin, are you? (I was only a kid when I watched it on the BBC.)

[JKH]

“It is the idea that the demand for goods is a function of the quantity of labour that households are actually able to sell. If households are rationed in the labour market, that will spillover and affect their demand in the output market. Because the amount of labour they are actually able to sell, and hence the income they will earn from wages plus non-wages, depends on demand. Which means that demand depends upon demand. Demand depends on itself. That was new, and interesting.”

Nick, I think that with respect to this issue, behavioral closure and accounting closure are inextricably linked in logic.

(Whether ex ante, ex post, continuous time, or apart from time)

However, if that makes little sense to you, perhaps this will:

It’s a 6 second video of the Keynesian Richard Nixon, who remembered what you remember:

https://www.youtube.com/watch?v=8qRZvlZZ0DY

And I’ll bet you remember that.

:)

[Edward Lambert]

Then if labor share fell say 8% and stayed low over time (capital share rises), you would anticipate no effect on full-employment aggregate demand from that?

Quotes from chapter 3...
"If in a potentially wealthy community the inducement to invest is weak, then, in spite of its potential wealth, the working of the principle of effective demand will compel it to reduce its actual output, until, in spite of its potential wealth, it has become so poor that its surplus over its consumption is sufficiently diminished to correspond to the weakness of the inducement to invest."
"Not only is the marginal propensity to consume[6] weaker in a wealthy community, but, owing to its accumulation of capital being already larger, the opportunities for further investment are less attractive unless the rate of interest falls at a sufficiently rapid rate;"

[Min]

Moi: "There is a time lag, so that demand(t1) depends upon demand(t0), where t1 is later than t0."

Nick Rowe: "it might lag. It might even lead. It depends on expectations."

No, there is no time travel. In fact, even if the predictions of future demand are accurate, there is a lag. Even if we **know** that the predictions are accurate. (But this is not the place for epistemology. :))

Nick Rowe: "You aren't old enough to remember Rowan and Martin, are you?"

I am old enough to remember when gov't worked. Before the rise of the Planet of the Pissants. ;)

[Roger Farmer]

Nick

Your first supply curve is the ONLY supply curve in the GT. All else is due to interpretations by later economists that try to make sense of it all (ineffectively in my view). The choice of wage units to measure GDP implies that, in a one good Cobb-Douglas economy, GDP and labor emoployed are one and the same thing. This is a clever device that allows us to concentrate on aggregate economic activity.

"There is absolutely nothing new on the supply-side in chapter 3 of the General Theory."

Not so. Although there is no 'theory' of aggregate supply that would satisfy a micro economist, that is not the same as your claim that there is nothing new. What is new is the assertion that Keynes will drop the classical second postulate. In other words: throw away the labor supply curve. Making sense of that statement is what my own work is all about.

http://www.rogerfarmer.com/NewWeb/PdfFiles/adas.pdf

I agree that the consumption function was a central element of the theory of demand. It is also the weak link. Attempts to reconcile short run cross section evidence and long run time series evidence on the value of the multiplier led to permanent income theory and the Ricardian equivalence debate. That debate is what we are all so heated up over right now. If the GT is nothing but the Keynesian multiplier, and if that theory is wrong, then what is the NY Times Keynesian left with?

We are left with the theory of supply; a denial of the classical second postulate. My work builds a foundation to the theory of aggregate supply that has a firm foundation in choice theory. That focuses debate on where it should be. What is the best way to restore effective demand?

Roger

[Scott Sumner]

Nick, Suppose someone argued that AD was (by definition) equal to PY. And that AD shocks caused business cycles because W is sticky and L = f(PY/W).

How would Keynes have objected to that model?

[Mark A. Sadowski]

Interestingly, I too have been making the argument elsewhere right now, along somewhat different lines, that there is nothing really new about the Keynesian aggregate supply function.

As to your second point:

"It is the demand-side that is new. It is the idea that the demand for goods is a function of the quantity of labour that households are actually able to sell."

I'm not even sure that this is really new. For example in The Theory of Interest (1930) Fisher essentially argues that demand is a function of production and the interest rate, and in turn production is a function of employment, so ultimately demand is a function of employment and the interest rate. So the idea that demand depends on employment was not really new either.

Or is there something about your argument that I am not comprehending?

[chris herbert]

First there is desire, then there is sex. Not the other way around. First there is demand, then there is the satisfaction of demand. Is the sequence unimportant?

[Nick Rowe]

Hi Roger:

"Your first supply curve is the ONLY supply curve in the GT."

Agreed. In the sense that there is that supply curve in the GT, and Keynes says the economy will always be *on* that supply curve.

But that supply curve is nothing new. If Keynes had said "According to my theory, the economy will always be on the production function, and will always satisfy W/P=MPL" the average "classical" economist would have rolled his eyes and replied "So, what else is new?"

And if Keynes had replied "What else is new is that the economy will be off the labour supply curve, except at full employment" they would have replied "sure, that is what happens when there is involuntary unemployment, but what causes that involuntary unemployment?".

And if Keynes had replied "A shortage of demand for goods!" I think they would have again replied "So what else is new?"

It is when Keynes lays out his theory of Aggregate Demand, that he starts saying something new.

Put it another way: Marginal Cost is just W/MPL. Saying "P=MC" is the same as saying "W/P=MPL". Keynes' AS function is just the MC curve, writ large, and expressed as a function of L, rather than Y, and with output measured in wage units. If Keynes had told Marshall: "Firms produce where P=MC, and MC measured in wage units is an increasing function of employment", Marshall would not have thought there was anything new.

But I have more sympathy for Keynes' consumption function than you have. Yes, demand for consumption will not just be related to *current* income, especially if the "period" is very short. But if people expect an extended recession, so that they will be unable to sell as much labour and output as they wish, not just now but in future, we should not be surprised if consumption demand drops (and investment demand too).

[Nick Rowe]

Scott: " Suppose someone argued that AD was (by definition) equal to PY. And that AD shocks caused business cycles because W is sticky and L = f(PY/W)."

I think that Keynes of the GT ch19 would respond: "Well, if you insist on defining AD as PY, I suppose you can, but why would a fall in P by 10% cause Yd to increase by 10%, given the world (and central banks) as we know it?" And would then have started talking about a fall in W and P causing a rise in M/W, and whether that would cause r to fall, and whether that would cause I to increase, and wouldn't it be easier just to increase M?

[Nick Rowe]

Mark: "For example in The Theory of Interest (1930) Fisher essentially argues that demand is a function of production and the interest rate, and in turn production is a function of employment, so ultimately demand is a function of employment and the interest rate."

I don't remember Irving Fisher saying that, but then it was a long time ago I read him. If he added that employment was sometimes constrained by demand for goods, rather than by supply of labour, then it does sound like the idea is there in Fisher. Though, it can always be said that nobody ever really comes up with anything new. There was always a precursor to anything (at least in economics).

[JW Mason]

. If the GT is nothing but the Keynesian multiplier, and if that theory is wrong, then what is the NY Times Keynesian left with?

They are left going back and deciding that the multiplier was right after all. :-)

I agree with you: If we reject the income-expenditure link, there is not much left of Keynesian macroeconomics.

On the question of precursors, Patinkin's "Anticipations of the General Theory" is the classic discussion, no? I found it -- and his essay on effective demand -- to be very helpful on all this. I think Patinkin would agree with Nick here. As would Keynes, since he thought that the consumption function, along with the liquidity preference theory of the interest rate were the two original contributions of the GT. Both being aspects of the problem of making decisions under conditions of fundamental uncertainty.

[Nick Rowe]

JW: I think I agree that his LP theory is new too. It is very un-"classical". But I have no idea if there were precursors. Maybe Wicksell? Did Lavington say something similar??? But even if there were precursors, integrating LP into a coherent theory of AD was an achievement.

[Donald Pretari]

From a very amateurish view:
1. There was a Usable Model for a Well-Functioning Economy.
2. It was not very useful in a situation of Debt-Deflation.
3. Unlike Knight, Simons, Fisher, & Viner, Keynes wanted to develop a Model that Comprehended both a Well-Functioning Economy and any Deviations from it. The others I mentioned had no problem with Different Models for Different Occasions. Policy wise, they were very close.
4. Keynes attempt cannot work because A Well-Functioning Economy & a Deviating Economy are not Symmetrical. They have Different Optics, as it were. Somewhere along the line, probably pretty early given how many Qualifiers there are on each page,Keynes realized it wasn't going to work out. The Book still has lots of Insights, but they apply to a Multi-Model Approach.
5. All of the Economists I mentioned were very aware that Advocating Debt in a Situation Partly Caused by Debt, was Counter-Intuitive, Paradoxical, Whatever. That should have been enough to forewarn anybody about attempting an All-Ecompassing Model.

The situation reminds me a bit of Quine's Indeterminacy of Translation. If you accept his premises, you spend a lot of time on puzzling problems that seem on the unsolvable side of things. If, however, you realize that Quine is a Behaviorist, a position you find faulty as it stands, you free yourself from these puzzling problems. It's an interesting exercise, but vitiated from the start by false assumptions.

[djb]

The most important part of chapter 4 [he means chapter 3 NR] is the end where keynes that in the macro if for a given period income is paid totally by sales only (consumption).... then income will be inadequate to produce those sales.....
Because people do not spend all their income on consumption....they save some

so the system would not be viable.

So investment is needed to be added to income

[Scott Sumner]

Nick, You said:

"I think that Keynes of the GT ch19 would respond: "Well, if you insist on defining AD as PY, I suppose you can, but why would a fall in P by 10% cause Yd to increase by 10%, given the world (and central banks) as we know it?""

That's just another way of saying "why assume that the AD curve doesn't change when the economy is hit by a real shock" And I agree, a real shock would probably cause the AD curve to change, given the way real world central banks behave. Indeed we just saw the transcripts of the Fed for 2008, and an adverse supply shock caused the central bank to shift AD to the left under any plausible definition of AD, not just P*Y. So I don't think that really answers my question. Your answer would equally lead to a rejection of any other arbitrary definition of AD. But perhaps that's how Keynes would have responded.

It seems to me you can engage in a quixotic quest for a AD definition that cleanly separates AS and AD shocks, or you can sensible give up and just talk about real shocks and nominal shocks, accepting that under most central bank reaction functions (except the correct one--NGDP targeting) a supply shock will lead to central bank reactions that shift the AD curve.

Yd seems like an artifact of the sticky price view of macro. If you switch to sticky wages and P*Y shocks, then there is no need for Yd. You have NGDP shocks and sticky wages. And that's all you need. What did Keynes tell us about the real world that this simpler approach does not.

[djb]

Meant ch 3

[Nick Rowe]

Scott: suppose you assume that P is perfectly flexible but W is sticky. You then have no need for Yd, because Y=Yd=Ys always, but you have a need for Ld. Because L=min{Ld,Ls}. And if Ld < Ls, then L=Ld. You could draw the Keynesian Cross diagram with L and Ld, rather than Y and Yd, on the axes. That would be a proper *Keynesian* Keynesian Cross, (rather than the Samuelsonian version).

Keynes' contribution? If you define AD as PY, then Keynes had a theory of what determined AD. And it was a different theory than the "classics". And it included the insight that AD depends on itself AD(AD), so that what we call an "AD curve" is really a semi-equilibrium condition, where each point on the curve is a fixed point, where AD(AD)=AD.

And this goes back to a point I made in some of my old posts where I say that existing NK models are fatally flawed because they just assume (everyone expects the economy eventually returns to) full employment, with no internal (model-consistent) justification for that assumption. To a first-order approximation, the marginal propensity to spend (out of permanent income) is one, so the Keynesian multiplier is infinite. This is where I both agree and disagree with Roger. I agree on the indeterminacy of equilibrium (and it's swept under the rug in NK models), but say it's coming from the demand-side, and not the supply-side. And it's all because NK models are models of monetary exchange economies without money. Without a real balance effect (in *some* sense, not necessarily Pigou), equilibrium is indeterminate (if it exists). Keynes knew this, and NK economists have totally buried Keynes' insight.

[Nick Rowe]

Edward: I just retrieved your comment from the spam filter.

You need to distinguish between the *marginal* propensity to consume and the *average* propensity to consume. The APC will be a declining function of Y (justifying what Keynes said in that quote) even if the MPC is a constant (so changes in the distribution of Y have no effect on C).

[Scott Sumner]

Nick, I agree with all that, and should have been more specific. I've also argued that Keynes's views on liquidity preference and AD formation is the actual contribution of the GT. I was thinking in terms of the impact of AD shocks on the economy. That's where I don't really see any new contribution.

As for the determination of AD, I think his model is very useful under a gold standard, much less so under fiat money. Under fiat money you model AD shocks by modeling central bank behavior.

Of course I agree with you about NK models.

[Tom Brown]

Nick, in layman's terms, what are the consequences of assuming P is flexible and W is sticky? What kinds of economists tend to think that and why?

What about vice versa?

What about thinking both are sticky?

I think I have an idea of who thinks neither are sticky: RBCers, right?

[Nick Rowe]

Scott: I totally agree, except maybe with the gold standard bit, which I don't really understand.

Tom: if either P or W is sticky, then AD (understood as P.Yd) matters for output Y and employment L, and it doesn't matter (much) which of the two is sticky.

If W is sticky, but P is not, then W/P rises in a recession, firms can sell as much Y as they want, but workers can't sell as much L as they want.

If P is sticky, but W is not, then W/P falls in a recession, workers can sell as much L as they want (but it will be a lot less than they would want to sell if W/P didn't fall), but firms can't sell as much Y as they want (and so Y falls, and Ld falls, so W falls and W/P falls until the involuntary unemployment is converted into voluntary unemployment).

If both are sticky then you get a bit of both.

If neither are sticky, then you either get RBC, or you get P and W fall without limit (if the AD curve is vertical in {P,Y} space).

[JW Mason]

I've liked the few posts a lot so it's useful to have Scott Sumner show up as a reminder of the things the Nick gets wrong. :-)

From my point of view, the problems come when you identify aggregate demand with (negative) excess demand for money, and when you imagine what the central bank does is setting the quantity of this money.

In a Walrasian world with only assets, it would be true that a reduction in desired spending on currently produced goods and services would necessarily imply an increase in desired spending on some nonproduced asset. But in a world if balance sheets, a fall in demand for goods does not require an increase in demand for anything at all. What the central bank is doing in this world is not changing the quantity of money, but changing the terms on which financial institutions trade off income and liquidity. In some cases, changes in aggregate demand are due to changes in the supply of credit which are subject to central bank influence. But in other cases they come from changes in desired spending by nonfinancial units. And again, a change in desired spending does not require a change in desired holdings of money or any other asset.

Of course I also agree with Nick's criticism of NK models. But I would solve the problem differently. I would prefer to accept that long run income growth really is indeterminate, and that spending decisions therefore cannot be made in an intertemporal framework.

[JW Mason]

In terms of stickiness -- the old Keynsian (or at least Leijonhufvudian) position is that it doesn't matter. This is a story about incomes, not about relative prices. I guess that would be Nick's last case in his comment above.

[Tom Brown]

Nick, thanks a lot! ... thanks for your input too JW (BTW, I found an error in my expression for K in the previous post & corrected it).

[Mark A. Sadowski]

Nick,
"If he added that employment was sometimes constrained by demand for goods, rather than by supply of labour, then it does sound like the idea is there in Fisher."

That idea is not really present in "The Theory of Interest", although that's where you will find his model of intertemporal consumption. According to Fisher, consumption is a function of present income, expected income and the rate of interest, among other things.

However, the idea that employment is sometimes constrained by the demand for goods seems implicit in "A Statistical Relation between Unemployment and Price Changes" (1926). There Fisher gives an explanation for his empirical observation that price changes precede employment changes:

"The principle underlying this relationship is, of course, familiar. It is that when the dollar is losing value, or in other words when the price level is rising, a business man finds his receipts rising as fast, on the average, as this general rise of prices, but not his expenses, because his expenses consist, to a large extent, of things which are contractually fixed, such as interest on bonds; or rent, which may be fixed for five, ten, or ninety-nine years; or salaries, which are often fixed for several years; or wages, which are fixed sometimes either by contract or custom, for at least a number of months. For this and other reasons, the rise in expenses is slower than the rise in receipts when inflation is in progress and the price level is rising or the dollar falling. The business man, therefore, finds that his profits increase. In fact, during such periods of rapid inflation, when profits increase because prices for receipts rise faster than expenses, we nickname the profit-taker the “profiteer”. Employment is then stimulated-for a time at least."

http://www.scribd.com/doc/8419994/I-Fisher-I-Discovered-the-Phillips-Curve

In short, increased nominal business receipts stimulate increases in employment, in part because wages and salaries are stickier than prices.

[Tom Brown]

Nick, O/T: Why did the demand for money go up tenfold here:
http://www.themoneyillusion.com/?p=26213#comment-320391
Mark A. Sadowski was "somewhat mystified" by your response and could not explain it.

[Nick Rowe]

Mark: interesting quote. But I don't see there the idea that a constraint on the sales of goods spillsover to affect the demand for those same goods.

Tom: suppose I am a tenant, and rent my house from a private landlord, because I like that particular house. Now suppose the government buys out my private landlord, but that makes no difference to me as tenant. My demand for renting privately-owned houses has dropped by 1 house, and my demand for renting government-owned housing has increased by 1 house.

[Tom Brown]

Nick thanks, that's what I imagined you'd say was the cause. I think David Glasner agrees with you (after a 90s analysis). I'll refer to that increased demand as a "substitution effect." Let's imagine that it's the CB which takes over banking just to make the situation crystal clear.

So the stock of "money" (can we call it MOA?) was increased (lost $1 reserves but gained $10 deposits), thus increasing supply by 10 fold, but there was no effect on long term price levels (long term neutrality of money) because the increased quantity of MOA was substituting for bank credit (bank credit is not MOA) thus there must have been a 10 fold increase in demand. Is that a fair description?

Note that in general MOA could go up or down, because on the CB balance sheet reserve-liabilities disappear, but customer deposit-liabilities are gained.

Can this substitution effect happen with any other non-MOA goods (other than bank credit)?

[Mark A. Sadowski]

Nick,
"interesting quote. But I don't see there the idea that a constraint on the sales of goods spillsover to affect the demand for those same goods."

Evidently I still don't understand your point. I'll have to meditate on this some more.

On the other hand with respect to Tom Brown's last comment, that's not the part I'm mystified by.

The monetary base consists of currency and commercial bank reserves. Broad money (usually) consists of currency and commercial bank deposits. It does not matter if the commercial bank is state owned or privately owned. So nationalizing a commercial banks has no effect at all on the size of the monetary base.

For example, most of the commercial banking system in France was nationalized in 1982-86. This made absolutely no difference in terms of the size of the monetary base or the amount of broad money. Moreover there are many examples of state owned commercial banks globally (e.g. China), but this has no effect at all on how monetary aggregates are defined.

[Tom Brown]

Mark, I understand your point about nationalization, but I'm not sure I understand what it is about Nick's explanation that you were mystified by.

It was because of feedback from Mark and Scott that I changed the example from "nationalization" to "CB takes over" and from talking about "base money" to talking about "MOA." Although we didn't use the term "MOA" I get the impression from the rest of the thread at Scott's that both Mark and Scott might accept that CB held customer bank deposit liabilities qualify as MOA.

[Tom Brown]

Summary of my question here:
http://brown-blog-5.blogspot.com/p/blog-page_2.html

[Mark A. Sadowski]

Tom Brown,
"Mark, I understand your point about nationalization, but I'm not sure I understand what it is about Nick's explanation that you were mystified by."

Tom, that *is* the part I'm mystified by. Nick appears to have said that nationalizing a commercial bank changes the size of the monetary base when it doesn't.

[Nick Rowe]

Mark: I think it's a semantic point. Do we define base money as monetary obligations of the central bank? If so, then if the central bank buys up a commercial bank, the monetary liabilities of that commercial bank now become base money. But nothing hinges on this. It doesn't matter.

[Mark A. Sadowski]

@Nick,
"I think it's a semantic point."

Almost everything is. :)

Tom's original hypothetical question essentially concerned what happens to the monetary base if commercial banks are nationalized, *not* what would happen to the monetary base if the central bank took over all commercial banks.

In the first case nothing happens and there are historical examples of this we can point to in order to verify that this is true.

In the second case, we're essentially talking about eliminating the division between central banking and commercial banking. Then bank reserves disappear and bank deposits become part of the monetary base. If bank reserves are larger than bank deposits then the monetary base increases. However, fundamentally nothing really changes.

But, to my knowledge, nothing like the second case has ever happened in real life.

@Tom,
The one useful thing that I hope that has come out of this discussion of hypotheticals is the realization that the inside money/outside money distinction that Gurley and Shaw popularized with "Money in a Theory of Finance" (1960), and which the Monetary Realists seem to be so enthralled with, is seriously flawed when it comes to talking about monetary aggregates.

[Nick Rowe]

Mark: "Tom's original hypothetical question essentially concerned what happens to the monetary base if commercial banks are nationalized, *not* what would happen to the monetary base if the central bank took over all commercial banks."

Aha! Then I misread it.

Yep on Gurley and Shaw. They are wrong on money as net wealth too. Pesek and Saving were better.

[Tom Brown]

Nick, would you say that "money" part of "long term neutrality of money" is MOA? So all else being equal (including no offsetting change in demand for MOA), if the quantity of MOA changes from M to F*M, then the long term average price level ought to change from P to F*P?

My hypothetical seemed like an odd case: a case where M changes but P doesn't. However, I do understand your explanation for it: demand for non-MOA bank deposits was exchanged for demand for MOA CB deposits (i.e. not all else was equal). Can any other non-MOA good (i.e. other than bank deposits) have this effect?

Also, what is the essential quality that differentiates a nationalized bank from the CB with regard to this hypothetical? It seems we could incrementally subsume the nationalized bank into the CB in tiny steps: first move their headquarters into the CB building. Then replace bank management w/ CB personal. Etc. At what point in this incremental CB-itization of the nationalized bank do its liabilities become MOA? Looking at the gov from a black box perspective (i.e. Tsy, CB, and nationalized bank all in the gov box) it's not clear to me what the critical step would be.

[Blue Aurora]

Although this response is a bit belated, I thought that regular followers on Nick Rowe's blog ought to read my comment on Roger Farmer's post on Keynes and aggregate supply.

http://rogerfarmerblog.blogspot.com/2014/03/did-keynes-have-theory-of-aggregate.html?showComment=1394261503457#c6544115229784712953