A Stable Spending Catechism

NGDP targeting, productivity norm, inflation targeting, unsustainable booms, supply shocks
Jules-Alexis Muenier, "La Leçon de catéchisme," https://commons.wikimedia.org/wiki/File:Jules-Alexis_Muenier_-_La_Le%C3%A7on_de_cat%C3%A9chisme.jpg

Although it has gained many converts since 2008, thanks especially to tireless crusading on its behalf by Scott Sumner, David Beckworth, and Lars Christensen, among other "Market Monetarists," the suggestion that the Fed ought to stabilize, not the inflation rate, or employment, but the growth rate of overall spending on goods and services, still strikes many people as odd, if not positively barmy.

Being, as it were, a sort of Market Monetarist avant la lettre  (for I first came to regard a stable level of overall spending as the sine qua none of a sound monetary regime while writing my dissertation ages ago), I naturally find the monetary policy credos of bona fide Market Monetarists as incontestably appealing as apple pie and baseball are to most full-blooded Americans.

My particular understanding of the case for stable spending is, nonetheless, mine alone, and as such somewhat distinct from that of my Market Monetarist brethren. So I thought I might venture, with all due humility, to try my hand at conveying that understanding to those curious but skeptical unbelievers among my cherished readers, by way of an imaginary exchange of questions and answers, where the questions are the unbelievers', and the answers are my own.


What exactly do you mean by "overall spending on goods and services"?

I mean the total quantity of money — of dollars, in our case — that consumers buyers hand over to sellers in exchange for finished or "final" goods and services (as opposed to intermediate goods or factors of production) over the course of some definite period of time.

A popular, rough measure of such spending is called "Nominal Gross Domestic Product," or Nominal GDP, which according to the IMF "measures the monetary value of final goods and services — that is, those that are bought by the final user — produced in a country in a given period of time (say a quarter or a year)." For 2016, total U.S. Nominal GDP was about $18.9 trillion, which was just under 3.6 percent higher than its level in 2015.

Strictly speaking, Nominal GDP measures the value of goods and services produced, rather than goods and services actually sold to consumers: the difference consists of changes to firms' inventories. A different statistic, "Final Sales to Domestic Purchasers," distinguishes actual sales from output. Final sales for 2016, at over $19.36 trillion, exceeded Nominal GDP, which means that firms' inventories were declining. The year-over-year growth rate of final sales, at 3.64 percent, was also slightly higher than that of Nominal GDP.

Why is it desirable that spending should be kept stable?

Because a stable level of overall spending, meaning in practice a level that grows at a modest and steady rate, helps to avoid recessions on one hand and unsustainable booms on the other.

What exactly do you mean by "recession"?

I mean a circumstance in which businesses as a whole are losing money — that is, taking in less than they spent, allowing for a "normal" return such as they might have earned by just investing the same amount in safe bonds. Alternatively, if you will allow me to oversimplify a bit for the sake of brevity, one might say that during a recession the "average" firm is losing money.

OK, I can see what you mean by that. But then, is the fact that a firm loses money necessarily to be regretted? Surely some firms ought to go out of business, or at least to cut back on it!

Of course some should! And that's not necessarily a problem. So long as the losses of some firms are matched by the extraordinary profits of others, there needn't be any unemployment, either of labor or of other resources. Instead, if the pattern persists, resources will move from the losing to the gaining businesses. And that's just what has to happen if there's to be as little waste as possible.

If, on the other hand, spending as a whole shrinks, there are more unprofitable firms that have to  dispense with workers than profitable ones seeking to hire more of them.

So the monetary authority would be justified in allowing the money stock to increase in the case where spending will shrink otherwise, but not to help firms that are in trouble despite stable spending growth?

That's correct. Resorting to money creation to help particular firms or industries, when the "average" firm isn't in trouble, only serves, first, to delay desirable changes in how inputs (the materials and labor businesses bring together in assembling their products) are employed and, eventually, to raise the prices of both those scarce inputs themselves and the product made using them.

But then why worry if the "average" firm loses money?

I repeat: for that to happen, it must be the case that the receipts of industry taken as a whole fall short of its expenses, or of its expenses plus a normal profit. In that case, few if any firms are inclined to hire more workers and to otherwise expand production, while others must curtail production and lay off workers. That spells general unemployment, that is, a recession.

Why should more money creation be useful in this case?

A decline in the "average" firm's revenues can only happen if people are spending less on goods and services, that is, if they decide to hold more money. Some expansion of the money supply can serve to make up for that extra demand for money, reviving spending just enough to make our average firm break even again. Individual firms might still fail; but then others would prosper. So long as total spending by the public on goods and services is stable, or increasing at a modest rate, firms as a whole — hence the average firm — can't be losing money.

So all that's needed to avoid hard times is to keep money flowing?

Whoa! Not so fast.  What I'm saying is that so long as there's no decline in overall spending, either absolutely or relative to some modest growth rate, there won't be recessions. But there can still be hard times: production could suffer because of wars, or bad harvests, or trade embargoes, or all sorts of other bad things. In economists' lingo, there might be "negative supply shocks." But there's nothing central bankers can do about them — indeed, no monetary system of any sort can do anything about them. Nothing that's likely to help, anyway.

OK. So suppose they just stick to worrying about spending. How can the authorities know whether they are allowing for sufficient money growth, that is, whether monetary policy is or isn't sufficiently "easy"? Must they keep track of how many firms fail?

Of course not! They just have to keep track of total spending itself.  To do that, they can look at the statistics like Nominal GDP, or Final Sales to Domestic Purchasers.

Sounds easy!

Well, don't get the wrong idea. It's still a tricky business. Those spending statistics don't come out very often. They also tend to get revised. So there's plenty of guesswork involved, and corresponding room for mistakes. But the principle is still sound — and even allowing for mistakes, a monetary policy that's based on wrong principles is bound to be worse.

OK. I get the idea that too little spending can be bad news. But can there really be such a thing as too much spending as well?

There sure can be! Just as too little spending means that the average firm suffers a loss, too much means that the average firm is making extraordinary profits, meaning profits well beyond what it needs to stay in business.

Is that bad?

It is, because it must lead to inflation in the long run — or an inflation of input prices at any rate — and may result in an unsustainable price boom in the short run.

How does too much spending cause a boom?

As I said, excessive spending causes the average firm, or firms as a whole, to enjoy extraordinary profits, at least for a time. Stock prices reflect firms' expected profitability, so it's natural for stock prices to rise when profits themselves rise unexpectedly, as will tend to happen if spending on goods and services itself accelerates unexpectedly.

And why should the boom not be sustainable?

Because, when firms are enjoying extraordinary profits, they endeavor to acquire more inputs so as to expand production. But when all, or most, firms are trying to do this, they only succeed in bidding-up input prices, because in normal times there are only so many inputs to go around. As input prices go up, so do firms' unit production costs. Their once extraordinary profits therefore cease to be so. When word of this hits the street, stock prices go back down as well.

O.K., I see why too much spending can cause an unsustainable boom. So, does that mean that the monetary authorities need to look out for such booms so as to put a stop to them?

Not at all! Monetary authorities are no better at telling whether a boom is sustainable or not than ordinary investors. If anything, they tend to be worse, in part because they have less to lose if they're wrong.

So what should the authorities do to prevent unsustainable booms?

I've already told you: they need only pay attention to total spending, that is, nominal GDP or some similar measure, making sure that it isn't growing too rapidly. So long as it doesn't, there won't be any unsustainable booms — at least, none for which monetary policy is to blame.

But how rapidly is "too rapidly"?

Now there's a good question! There's room for expert disagreement on this point, but the disagreements aren't dramatic ones. Some would have spending grow enough to result in an average inflation rate of 2 percent, which is the rate most monetary authorities favor. That means letting spending grow at a rate equal to the economy's long-run real growth rate, plus another two percentage points, or at around 4-5 percent per year.

Yours truly, on the other hand, has argued that the most stable arrangement is one that lets spending grow only enough to compensate for growth in the economy's workforce and capital stock, which in the U.S. today would mean having it grow at a rate of 2 or 3 percent per year. Since having spending grow at a steady rate means letting the inflation rate mirror the rate at which the economy's overall productivity improves, such a modest rate of spending growth would actually result in mild deflation much of the time. Since the deflation would reflect falling production costs, it wouldn't be a bad thing.

Of course there are those who hold other views, most of which (though not all) lie somewhere between the two I've just described.

Hmm. You are starting to get fancy. Why not just have the monetary authorities target a particular rate of inflation, or, if one prefers, deflation, instead of targeting spending?

Because stable spending is what's required to keep firms from either generally losing money or generally enjoying unsustainable profits. A stable inflation (or deflation) rate isn't generally consistent with that outcome.

Why not?

Because an economy's inflation rate depends on two things. The first is how much people spend. The second is how many real goods and services firms are able to supply using the inputs they buy. If the rate of growth of spending is itself stable, the inflation can still vary as productivity (the output produced from any given amount of inputs) varies. So long as productivity fluctuates, a stable flow of spending requires a fluctuating rate of inflation. Keeping inflation stable, on the other hand, would mean letting spending fluctuate. And we've already seen that fluctuations in spending aren't consistent with avoiding unsustainable booms and busts.

I suppose. But just how much harm could possibly come from, say, sticking to a constant rate of inflation?

Plenty, actually. Let me give you a for instance.  Remember what I said about negative supply shocks — you know,  wars and bad harvests and all that?

Yeah, sure. Causes of hard times that monetary policy can't fix.

Exactly. But while no monetary policy or regime can undo the harm done by a negative supply shock, the wrong sort of policy can make things worse. Suppose, for example, that a country becomes the victim of a blockade. Because fewer goods are available, people are bound to be worse off.  If spending stays stable, the shortage of goods will also mean higher prices all around.

Right! So why not at least take steps to prevent the inflation? People are already suffering enough!

No, no no!  You've got it all wrong. The inflation in this case is just a reflection of the fact that goods are scarcer than usual. The economy is only suffering from one "bad" thing — not two! Suppose the authorities decided to "do something" about the inflation. What would it be? The answer is that they'd have to tighten money, and limit spending, to keep prices from going up. In other words, to make up for the greater scarcity of goods, they'd be making money more scarce as well! So instead of just finding that their incomes don't buy as many goods as before, now people get fewer goods and less income to buy them with!

So stable spending really does beat stable inflation. You know, I think I'm starting to get it! 

Kudos to you!

But you still haven't explained which view regarding just how quickly spending ought to grow is the correct one.

That's true. But really, the actual spending growth rate matters a lot less than having spending grow at some steady and predictable rate. For once people and businesses become accustomed to that rate, the mere fact that they know and can predict it will suffice to rule out serious business cycles, at least so far as monetary policy can do that. So why don't we leave it at that for now. Getting people to see why responsible monetary policy is fundamentally about keeping spending stable is hard enough as it is, without trying to convince them that a spending growth rate of 4 percent is a whole lot better than one of 2 percent!

O.K., I get that. But all this talk about the right monetary "policy" makes me uncomfortable. Why entrust the management of money to a bunch of bureaucrats in the first place?

Congratulations! By asking that intelligent question, you have earned a place at the head of the class!

The truth is that, so long as we rely on discretion-wielding central bankers to manage national money stocks, we're unlikely to witness the sort of stable spending that I believe is most consistent with overall macroeconomic stability. There are all sorts of reasons for that, some stemming from political pressures, others having their source in central bankers' limited knowledge.

I personally think we can do better. One option is to impose a monetary rule or mandate or both giving central bankers no option save that of maintaining a stable flow of spending. That would at least stop them from undermining stability by pursuing other goals. But I believe we can do better than that, by reproducing some features of the remarkably stable "free banking" systems that blossomed in several places in the past, and especially those features of free banking systems that served to automatically stabilize spending. For the time being, though, we are stuck with monetary systems to which bureaucrats hold the reins, so we must do whatever we can to make the best of a bad bargain.

In any event, whatever monetary system one prefers, it remains the case that, if that system is to work reasonably well — if it's going to manage money better than central bankers have — it will have to succeed somehow in achieving and maintaining a stable flow of spending.

  • M. Camp

    I'm still unclear on the meaning of total spending. Is it "consumer" spending[1], or total final goods spending[2][3] (including both consumer goods and production plant and equipment?

    = = = = =

    My emphasis:

    [1] "I mean the total quantity of money — of dollars, in our case —
    that **consumers** hand over to sellers in exchange for finished or "final"
    goods and services (as opposed to intermediate goods or…"

    [2] "…**factors of production**]) over the course of some definite period of time.

    [3] The most popular measure of such spending is called "**Nominal Gross
    Domestic Product**…"

    • George Selgin

      Thanks for these observations. "Consumers" was admittedly sloppy on my part. I ought to have said "buyers," for it is indeed spending on all final goods that I believe must be kept stable. As for NGDP, although it is indeed "the most popular" rough-and-ready measure of such spending, I do not mean to deny that it is not an accurate measure.

      • M. Camp

        Thanks much. I was *almost* sure you meant "buyers": your aim* is to reduce (by inference) ALL price falsification, not just the consumer sort.

        * Monetary Policy Primer Part I: "shortages and surpluses of money hamper the correct setting of individual prices…".

  • Ray Lopez

    Money is neutral, short term and long, but George Selgin is right: the less central control we have of money, the better off we are as a society (less bureaucrats salaries to pay). Anyway with digital money now (physical money being less than 8% of GDP), the money supply is becoming irrelevant. And the Fed follows the market. The few times they've tried to buck the market they had to back down. Even in Fx markets, clearly manipulated by central banks, the central bankers follow the markets (and that includes the famous Plaza Accord, just look at the charts).

    • Hannibal Smith

      The Fed doesn't follow the market; the market anticipates and discounts the Fed's action. Correlation is not causation.

      But fixed exchange rates have never worked. The market is always right in that case.

      • Ray Lopez

        Because of human nature we have booms and busts. You can even mathematically model booms and busts, see the work of Brian Arthur (1983) keyword "Generalized Urn Problem". As for 'the market anticipates…the Fed' you can't prove that, though it may be true, so scientifically, it's false. As for free wildcat banking is a huge social negative, you can also say that for the monopolized central bankers. George Selgin for President (of a decentralized Fed)!

        • Hannibal Smith

          You could probably "prove" it by lagging T-Bill vs FFR changes. There's no transmission mechanism from the Fed to the real economy, so the market has to discount potential policy changes. If you track the Fed Futures, they're not at all that terribly accurate until nearly the event date.

          Central banking isn't a "huge" social negative — you're smoking the fantasy crackpipe that Ivory Tower academics like Selgin believe in. I suggest learning about operational reality and you'll discover that the Fed is the Wizard of Oz. It's anarchronistic Jekyll-Creature-conspiracy-theory-mumbo-jumbo to believe contrarywise… that fits certain political ideologies.

          • Ray Lopez

            Thanks but your narrative seemingly presupposes strong money non-neutrality, or traders belief in such (which admittedly perhaps is strong–day traders also believe in nonsense like 'head-and-shoulders' formations and 'double tops'), which the evidence from what I've seen (Bernanke et al, FAVAR paper 2002, showing a small 3.2% to 13.2% effect of Fed policy shocks on the real economy) is weak. I agree with you that the Fed = Oz. Ergo, it follows any Granger-causation will also be weak, regardless of which variable is leading or lagged.

          • Hannibal Smith

            That sounds a lot like that branch of Rational Expectations Economics or whatever it is called. But I'm not sure its completely realistic to just say "Well, the market will just negate the effects of any Fed policy." because obviously a lot of investors are always caught unawares with their pants down, otherwise they wouldn't act after the event to produce such "shocks". But since the Fed only controls the FFR/DR, there's no way that short-term T-Bill rates can rise to match the new FFR rate as they do except via the secondary markets (I'm ignoring the triumvirate "Treasury auction" here as I can't remember if that is more MMT B.S. or operational reality) or private contracts pegging on the rate.

            One thing I do agree about is that the Fed is always fighting the last war and is always lagging the market, but I suspect that's more of a bureaucratic-political consequence than by intent. A Politburo didn't work very well in the ex-USSR; we shouldn't expect it to work any better in the USA. Though as you've indicated that you understand, the Wizard of Oz effect is indeed primary and not any actual transmission mechanism. My ongoing annoyance is that the majority of everyone interested in this field keeps treating the former as if it were actually valid in reality! It's fine up to a certain point to claim the "big bad central bank controls the entire money supply" for purposes of promoting what is believed to be private money alternatives, but such pundits are not going to win credibility or arguments with operational reality pragmatists by trafficking in hallucinatory B.S. (are you listening, Professor Selgin???).

            BTW, chart patterns aren't nonsense… they merely reflect demarcation lines draw among significant emotional extreme points (i.e. lack of buyers or sellers), hence there's memory effects built into the pricing mechanism. Nothing controversial there; you just would expect chart patterns to be higher probability above a coin toss from all the marketing fiction. I've never been impressed that any pattern has ever lived up to the mystique and I've been examining them for two decades. I suspect they worked much better when humans were actually doing the super majority of trading (i.e. pre-1980's) instead of algorithms that aren't limited to simple linear visual constructs. There's also been a decades long reversal of breakouts working in favor of mean reversion which doesn't bolster their case either.

            P.S. A new cryptocurrency with built-in "monetary policy" to keep it value stable is debutting in a few months.

          • Ray Lopez

            @H.S. – it's not Rational Expectations, but the exact opposite: agent-based theory. R.E. assumes a 'representative man' so you never get shocks or surprises (everything, on average, works out) while agent-based 'chaotic' theory says you can get short-term wild swings (even if on average things might work out the same, in the long run). But–this is key–even back in the late 1980s (Baxter & Stockman, 1989, not David Stockman but another) found that wild nominal price swings on the Forex market did NOT produce any real effects. More concrete example: everybody "knows" inflation is "bad" but in fact per capita growth rates during moderate to high inflation (1970s) were about the same as in the 1980s (with low inflation). That said, during "hyperinflation" bad things happen (Zimbabwe) but most of the time people anticipated changes and react to them, either as per R.E. (on average) or even like agents-theory anticipating changes and usually getting it right, but even when they don't nothing bad seems to happen.

            Chart rules / Oz effect – are usually at best short term (day traders) even algo traders (the chart rules may be programmed into the bots), and as you imply, once these short term effects become prominent they seem to disappear (not impressed as you say). It's fairly well known stock markets, though nearly a random walk, have 'memory' as you describe*, even over long periods of time like quarters not just milliseconds, but trying to profit off these 'charts' is hard to do (or a lot of people would be rich). Even uber-math rules trader (quant arbitrageur) Jim Simon would agree.

            * from Cuthbertson's textbook on finance: "There is some evidence that active trading strategies (market timing) based on the predictions from regression equations, may result in returns corrected for (ex-post) risk and dealing costs, which exceed those for a passive strategy but only over certain data periods"

          • Hannibal Smith

            Yes, virtually everyone in the 70's adapted to the higher inflation and didn't suffer in real terms except for about a bit over 1/4ths of the lower class. However, next time there is runaway "stagflation" and mainstream doom porn about Bretton Woods or the USD imploding blah blah blah, we may not have a wage price spiral locked in via employment contracts, so the outcome could be a lot worse for the asset-less poor. Hopefully by that time technological unemployment will be so severe we already have Citizen's Dividend implemented.

            Day trading is too short term for chart patterns. Market structural anomalies dominate intraday with liquid and illiquid periods and other repeatable occurences. Chart patterns take weeks to months to form as they're supposed to be (or used to be) representative of longer term investor's emotional outlier points. If you want to see statistics on chart patterns, go to Bulkowski's website. Only about one of the entire motley collection has impressive stats and yet in actual trading, it hasn't been impressive so far. I have corroborating evidence from two other sources that quantify chart patterns with the expected gains and you might as well fall asleep. "That's all? Zzzzz."

      • George Selgin

        "But fixed exchange rates have never worked. The market is always right in that case

        They did work, during the classical gold standard era, which lasted several decades. Admittedly the sort of "fixing" involved then was a far cry from the sort involved when central banks fix exchange rates today. See on this my Cato paper on "Law, Legislation, and the Gold Standard."

  • Another question for the catechism, since you specifically mention final goods: what difference does it make targeting spending on final goods rather than all goods? Why shouldn't changes in the flow of spending on intermediate goods also be compensated by changes in the money stock?

    • M. Camp

      I think it must be the same reason intermediate goods aren't counted in GDP, not wanting to double-count spending.

      I think of it this way. If there are two processes needed to produce a final good from resources and labor, and both are done in the same factory, then the spending is counted as the final sale to the purchaser. If the owner sells off the half of the business that performs the first process, then there has been no change in production, no change in incomes, and yet if we count intermediate goods, "production" has doubled, or so. So this is a false accounting.

      • Great thought experiment, thanks. So you're saying that vertical integration and disintegration, as pure organizational changes, shouldn't affect a monetary target?

        It's not obvious to me that it definitely shouldn't, since in the case of disintegration money is being used where internal accounting was being used before, and I don't see why this would necessarily be a macro-neutral event if it were large enough.

        I suppose you could point to transfers of money from one department to another as a kind of reductio ad absurdum. But informational issues notwithstanding, in principle I don't see why that should look any different from the transfer of money from myself to the grocery store from the monetary authority's perspective. If you draw a boundary around a set of agents, you can always say that transfers within that boundary don't affect spending outside of it and shouldn't count toward the total. So what makes "a productive process" a better boundary than "an organization", or something even more arbitrary like "the grocery store plus me" – and what criteria do you use to make that determination? Is it purely the ease of constructing an aggregate?

        • M. Camp

          "So you're saying that vertical integration and disintegration, as pure organizational changes, shouldn't affect a monetary target?"

          I'm wondering if that is what Prof. Selgin is saying.

          It doesn't follow from

          — Degree of vertical organizational integration must necessarily excluded from a production measure


          — it should be excluded from a monetary target.

          But it is one thing that he MIGHT be saying. I'm hoping to find out.

          Three money flows to Factors do not do not depend on degree of vertical organizational integration:
          — Labor
          — Landowners
          — Resource Owners .

          But once a supplier has his payment, a fraction of the eventual total of the return to capital (profit) has already been monetized, and he will likely distribute it to Households as dividends, earlier than the same flow would have occurred if his part of production were integrated.

    • George Selgin

      If final goods sellers earn normal profits on the whole, then that suffices to take care of the intermediate goods. On the other hand, broader spending measures can include financial flows such as can vary independently of spending on final goods and services. Stabilizing the broader measures can therefore mean destabilizing the one that ought to be stable.

  • mlouis

    The "two wrongs" can easily happen under your construct. Let's say real potential output growth falls to very low levels. Under your construct not only is society struggling with low real growth but also with inflation. This is a recipe for social unrest. A monetary policy rule is only as good as its ability to not be voted out by an unhappy population.

    • George Selgin

      I pose you a question: a war has destroyed the productive capacity of an economy. Under my preferred regime, with constant spending (Py), P goes up as y goes down. Now, you tell me how your preferred monetary system can make things better, bearing in mind the identity MV=Py. You can raise M, and thus make P rise further. Or you can reduce it, to keep P stable. Which of these options is it that you regard as better at keeping the population happier than my own preferred regime would?

      • Hannibal Smith

        Really, you're using a hyperinflation scenario and the bogus quantity theory of money (QTOM) to get your point across? LOL!

        Be that it may, aren't you essentially talking about automatic stabilizers here with your "constant spending"? If so, then I'm in agreement that it is a good idea to implement for "a happy population" so long as you recognize that it is accomplished via fiscal policy, not monetary. Under our present monetary system, M (or V or whatever other variable is deludedly "fixed" in the QTOM) already drops during adverse economic scenarios as people stop borrowing and creating private money. Monetary policy cannot offset a drop in y unless you give to the Fed illegal fiscal policy powers like, say, the Reserve Bank of Zimbabwe. And even then I have my serious doubts since Japan tried "constant spending" for decades to no avail (and the EU more recently), but sure has a lot of malinvested "bridges to nowhere" infrastructure no one really uses or needs.

        However, "constant spending" doesn't do anything about lower incomes directly which is where the lower M (or V, etc.) is derived from unless that "constant spending" is focused on transfer payments. So is the population ultimately unhappier to be employed with declining income (i.e. Japan) or are they unhappier to be out of a job with no income at all (i.e. USA)?

      • mlouis

        I don't think we can say which is better without digging into details. The way we increase M today is via lower interest rates: essentially increasing the value of financial assets to the point that enough economic activity is created to stabilize P. So if a supply shock raises the costs of a broad cross-section of the population, we would respond by increasing the wealth of those with financial assets (highly concentrated as we know) and decreasing real wages (since wages are sticky). This might stabilize P, but it seems to me like it would create a highly unstable social dynamic whereby every dip in spending is countered by pushing up the wealth:gdp ratio and widening wealth inequality. That will go on until society revolts. Hmm, feels a lot like our current system.

  • Andrew_FL

    The "Market" in "Market Monetarism" comes from the idea of using market forecasts of the nominal level of spending as the "target" of monetary policy. Stablizing the spending stream is actually of course as you know a much older idea. Implicit in most forms of monetarism or proto-monetarism but not unique to them either.

  • Mattyoung

    'OK, I can see that', says Eric Cartman of South Park.
    The method attempts to prevent bunching up such that firms losses and gains can always be separated and marked to market, it prevents jamming in the queues for a monopoly central banker. Fair pricing maintained, price variation is a fixed relationship with NGDP growth because of stable queuing effects..

    It works, but the uncertainty of money growth becomes a prime pricing wedge. We cannot price anymore accurately than we can measure NGDP growth (on a relative basis) . Trying to price beyond that accuracy results in currency risk, currency risk is priced and accumulates, and has to be paid for. But 2% NGDP growth works when the economy generally grows about 2%, then all the bounds on queues are about the same, the currency risk limited, and shared.

  • tmtisfree

    "Some expansion of the money supply can serve to make up for that extra
    demand for money, reviving spending just enough to make our average firm
    break even again."

    There seems to be some kind magic I don't understand in this sentence. People do not want to spend on goods and service, OK. But how this translate for them demanding for money? Logically, if people don't spend on goods, they already keep more money. How can that be they want some more which they will not spend anyway (not to speak of the decreasing purchasing power of the operation)? Make no sense to me.

    PS: "avant le letter": the correct one in French is "avant la lettre"

    • George Selgin

      tmtisfree, first of all, thanks for spotting the botched French, which I've corrected. I'm not sure how I managed to screw that up, 'cause I was as surprised to see it as you must have been!

      Regarding the more substantive point you make: when people spend less on goods and services, they are implicitly increasing their demand for real money balances. The demand for money is nothing other than a name to hold on to rather than part with monetary receipts that comes our way. I discuss the matter further here: https://www.alt-m.org/2016/04/28/monetary-policy-primer-part-2-demand-money/

      • tmtisfree

        Thank you for pointing to your part2 primer. The "demand for money" appears to be defined (by economists at least) as a desire to keep a surplus of money by not spending it. I'm fine with that.

        What I have trouble with, are the following assumptions that this "demand" has to be met by 1) a (net?) addition of some money somewhere in the system sometime in the future (uncertainty monster!) 2) to supposedly revive spending (for the average firm to make some profit, keep unemployment low, etc.).

        1) It is far from evident why people keeping money by not buying some goods/services should somehow imply people wanting quantitatively more money which has then to be supplied. There is a gap in logic here that this ex-scientist has problems to fill: this looks like as a jump to a conclusion. Have we data, mechanisms and/or adequate explanation(s)/justification(s) to accept it?

        2) Nevertheless accepting 1): adding some money to compensate some non-spending will decrease the money purchasing power (PP), negating the very purpose (or at least a part of it) of people keeping money: how people will react to such a scheme that goes against their will? By keeping a bit more money to try to achieve their initial goal (and regain some of the PP they lost) or to finally spend it (before PP decreases too much or whatever the reason)? How do we know for sure that the second assumption is the only one possible? What are the real world response(s) to injection(s) of money in general? More spending, less spending? Do we have "experiments" showing this or that response is better than what people have decided at first?

        I can accept that to deal with complicated matter, some models are possibly useful and I have no problem accepting assumptions as working hypothesis. With these in mind the monetarist position and given explanations are understandable, which your well-written and argued posts usually illuminate very clearly. But are these position and explanations justifiable with regards to the given assumptions?

        I have the feeling that many are taken for granted. Scientifically, assumptions have to be assessed against the real world because it is the only proper benchmark ; we have to find justifications to use them in order to advance our understanding. But perhaps I am too skeptical (or worst, naive).

        • George Selgin

          "It is far from evident why people keeping money by not buying some goods/services should somehow imply people wanting quantitatively more money which has then to be supplied. There is a gap in logic here that
          this ex-scientist has problems to fill: this looks like as a jump to a conclusion. Have we data, mechanisms and/or adequate explanation(s)/justification(s) to accept it?"

          Filling the gap is a matter of observing that when the increase in money demand is general, while the nominal stock is fixed,then it is literally impossible for anyone to add to his or her money holdings except by depriving someone else of the ability to do so. In the long-run, downward movements in prices will restore equilibrium, by increasing the purchasing power of each existing unit of money, thereby also increase the "real" quantity of money to accommodate the increased "real" demand for it. But such an adjustment may take time, during which the shortage of money translates into unsold goods and services, hence into unemployment.

          I hope that helps, and I apologize for the delay in replying.

  • Andrew_FL

    It occurs to me that one particular Market Monetarist view which is not addressed here is Level Targeting, the idea that the Central Bank should try to retroactively correct its mistakes in targeting spending growth by pushing spending back up to trend. Where would you say you stand on that?

    • George Selgin

      I agree with level targeting. I tried here to make my language here consistent with that view, though without having wished to venture into a discussion of the distinction you raise, as I don't think that would be in keeping with the spirit of this post.

  • Ralph Musgrave

    I find the hypothetical blockade mentioned above very unrealistic: Trump
    would just drop a MOAB on each blockading ship. Or maybe something smaller than
    21,000 lb would do the job?

    But seriously, to cut spending in reaction to inflation stemming from a
    blockade would arguably not do any harm, and for the following reasons. Suppose
    to keep things simple inflation is directly related the amount by which
    aggregate spending exceeds the aggregate ability of the domestic economy to
    supply: e.g. if the former exceeds the latter by 10%, then inflation is 10%.
    Suppose a blockade is imposed which produces, for the sake of argument, the
    latter 10% disparity and 10% inflation.

    Government then cuts spending by 5%, which would cut inflation to 5%.
    That, however, would not cut real GDP because GDP is constrained by supply, not
    demand. Net result: GDP stays the same, but there’d be 5% inflation instead of
    10%. Good result I’d say.

    But that is a minor criticism of George’s basic point, because the
    blockade scenario is an unrealistic one. Possibly a more realistic scenario is
    a re-play of the 1970s when inflation rose because of excessive wage demand (at
    least that was the case in the UK I think). In that scenario I’d advocate what
    Paul Volker did: deliberately constrain demand and force unemployment up till
    inflation is under control.

  • Spencer Hall

    That prescription is quite wrong. And it is axiomatic. If the Fed targets N-gDp, inflation will ultimately consume us. The error is simple, and universally mis-understood. Commercial banks do not loan
    out savings (existing deposits). If one examines the exogenous inputs to the expansion of new money and credit, then it becomes obvious, “outside” factors are peripheral to the expansion of new money (as anyone who can tell a debit from a credit would already know), that the commercial banks are credit creators, not credit transmitters, always creating new money whenever they lend/invest. Apparently economists learned nothing about the goldsmiths, and their issuance of gold-certificates.

    • George Selgin

      Actually, Spencer, I've written at some length about the goldsmith bankers, and they never issued gold "certificates," as you claim. See "Those Dishonest Goldsmiths," Financial History Review. As for your statement, "If the Fed targets N-gDp, inflation will ultimately consume us," since it doesn't even specify a target growth rate, it makes no sense at all. As regards the target rates of 2-4% or so that I refer to in my post, it is simply not true. 2% growth would often be deflationary; 4% would usually mean mild inflation.

      • Spencer Hall

        That wasn’t even my point. You don't understand macro (along with ALL other economists). And note aside: the Goldsmiths in in seventh century London issued a certificates of deposit which possessed a wide acceptability that a credit instrument was created which could be said to embody the characteristics of paper money.

        Lending by the DFIs is inflationary. Lending by the NBFIs is non-inflationary, ceteris paribus. So, if you understood money and central banking, you would see the error in your thought process. You and Larry Summers don’t understand how secular strangulation (chronically deficient AD), and stagflation (business stagnation accompanied by inflation), are generated (because you don’t know a credit from a debit).

        Take the “Marshmallow Test”: (1) banks create new money, and incongruously (2) banks loan out the savings that are placed with them.

        F. Scott Fitzgerald: “The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.”

        You have to retain the cognitive dissonance capacity, like Walter Isaacson described Albert Einstein’s ability: to hold two thoughts in your mind simultaneously
        – “to be puzzled when they conflicted, and to marvel when he could smell an
        underlying unity”.

        John Maynard Keynes couldn’t do it. Bankrupt u Bernanke couldn’t do it” “Money is fungible. One dollar is like any other”. LOL!

        -Michel de Nostredame

        • Spencer Hall

          If you understood money and central banking, you would know why, and how the U.S. Golden Era in economics ended. And you could predict the path of the economy and interest rates!

          • Spencer Hall

            And there was Ming dynasty, 1368 to 1399 A.D., Chinese paper

    • George Selgin

      Spencer, the more you prattle on about how nobody (apart from yourself) knows any monetary or macro economics, the more you invite sensible people to embrace the alternative, more parsimonious theory that you yourself know neither. Claiming that you can "predict the path of the economy and interest rates" makes that alternative theory all the more appealing.

  • Philon

    What does 'final' mean? In general, goods and services are bought to produce results. To the extent that the purchaser's intention is to produce a further good or service *for sale*, the original good/service is intermediate; to the extent that the intention is something else, the good/service is final. Let's look at some examples.

    1.I buy a pack of paper, which I intend to use to write letters to my mother. Although I plan to use the paper to produce a further product–letters–I am still the final user, since I do not plan to sell the letters: I expect to produce them simply for the satisfaction of my mother and, perhaps, also of myself.

    2.I buy a pack of paper, which I plan to use to write a novel, which I hope to get published and copies of which I want to sell to the public. Here the paper is an intermediate good, so it does not figure in "Final Sales to Domestic Purchasers" (FSDP); only the revenue (if any) from sales of my novel counts. (But, as in the first case above, writing the novel may have given me some intrinsic satisfaction, so my use of the paper may have had mixed motives, part commercial, part personal; thus the paper may be partly final, partly intermediate, and I may be in part the final user, in part not.)

    3.I buy a new jacket, intending to wear it. But when my roommate sees it on the hanger he is so taken with it that he offers to buy it from me, offering a price above what I paid. So, never having worn it, I sell it to him. But I, not he, am the final user, for I bought the jacket not intending to resell it. If I had bought it intending to resell it (at a higher price, of course), my purchase would not count in FSDP, only my subsequent sale (if any; if I find that the jacket is, after all, unsaleable, and so give up my original intention and start wearing it myself, the increment to FSDP will be zero). If I had planned to wear it a few times and then resell it, my motive would have been partly personal, partly commercial, and I would be in part the final user, in part not. (I would want in part to consume the jacket, in part to use it as an input to the process of producing a second-hand jacket for sale.)

    4.I buy a new jacket, to be worn to work so that I will make a good impression when soliciting or advising clients, and to be worn also on social occasions to enhance my prestige. In the former use it is an intermediate good, used in the production of advisory services; in the latter use it is a final good (arguably). Part of the price of the jacket should be counted in FSDP, part should not. Here I have treated my desire to look good for clients and not my desire to look good in general social settings as having a commercial nature; but perhaps the latter is not *wholly* non-commercial, contrary to my treatment.

    In practice many of these niceties will be unmeasurable, and our statistics will no more than crude approximations.

  • Benjamin Cole

    Good post.

    I think the downsides of a central bank shooting for a hot economy—in a world of global supply gluts—are rather slim now.

    Also I have this plea: I like free enterprise, for both practical and philosophical reasons. Freedom, prosperity, etc.

    Okay, so we want to public to like free enterprise. They vote after all.

    A nation with "labor shortages" is a happy nation. Voters perceive the system works for them.

    The Fed shoots to keep at least one of of every 20 people who want to work instead unemployed. We lose the output too, and one could say bad lifestyles are strengthened when people are unemployed.

    This level of unemployment also results in a chronic situation of more people looking for work than job openings. The Fed seems satisfied when there are 1.4 people looking for work for every job opening. Oh how nice.

    But why not two job openings for every person looking for work? Who decides which level is right?

    Add on the reality open borders for illegal workers, and property zoning to keep housing supplies tight.

    So…how should the public vote? I am surpassed Sanders did not win in a landslide. As it was, the Sanders-Trump combined vote is something to think about,

    PS Japan has more job openings than unemployed, and deflation.

    In the oil boom. inflation in North Dakota and Texas was at national averages.

    We are keeping people unemployed in some sort of atavistic "tilt the playing field against working people" paradigm.

    PS yes get rid of minimum wage—and also the routine criminalization of push-cart vending, And end property zoning, including the zoning of land as "retail."

    • George Selgin

      Benjamin, I have always thought it prudent to limit myself to opinions regarding desirable monetary institutions and policies, while distinguishing between what we can hope to achieve by improving those, and what cannot achieve that way, but may wish to achieve by other means. As you can see, even sticking to that narrow agenda hasn't spared me from a barrage of criticisms and insults–some from people who think I'm not allowing for sufficient monetary accommodation, but most, alas, from those who regard any sort of monetary expansion as a bad thing.

      So, I'm reluctant to offer an opinion–that is, one for public display–regarding the ideal level of "natural" (frictional or structural) unemployment, though I'm quite certain that, whatever that level is, a monetary policy that leads to anything other than stable spending will not improve upon it for long, and will tend eventually to make it worse than it is. I doubt any candidate for office concerns himself or herself with such matters, or that any would resist favoring bad monetary policy were it capable of gaining enough votes. But that, of course, is a matter of what the voters understand, when is the margin I'm determined to influence, however meagerly or unsuccessfully.

  • jandr0

    [If, on the other hand, spending as a whole shrinks, there are more unprofitable firms that have to dispense with workers than profitable ones seeking to hire more of them.]

    Good. If they are unprofitable, then they do not deserve to be in business. Let their physical capital be re-appropriated and put to better use by other entrepreneurs. Let those entrepreneurs hire those "dispensed" workers at their marginal value. The economy will sort it out and recover, provided no delusional technocrat thinks he (or she) knows better and interferes.

    But voila! It is precisely here that our caped crusader and resident tecnocratic superhero, Selgin swoops in from the sky to peddle his mantra, since he believes he has the "answer" to save us from ourselves – if we would just believe in him.

    From this point on Selgin the technocrat piles interference upon interference to showcase his "technological prowess" (he just can't help himself) and "direct" the economy by "saving" unprofitable firms by mucking around with the money supply.

    Fiscal Keynes. Running a deficit becomes never-ending deficits. But that's fine, we deserve and can have "guns and butter" simultaneously. Yeah right. No 70's stagnation, eh?

    Plucking Friedman. Just grow with economy. Don't worry, there won't be things like Greenspan Puts, dot com crashes or housing bubbles.

    Pretender to the "I have the answer" throne:

    Stable Spending Selgin. Will fix all the above errors. Who knows where this will lead!?

    I suggest Selgin should write Hayek's term, "fatal conceit," on a big board and hit himself over the head with it every morning when he wakes up.

    Sorry George. I've listened to your talks. Read your book. You come across as quite a nice guy. But you want to mess around with sound money (probably for what you regard as justifiable good intentions and oh-so-many self-rationalising noble reasons).

    In the end, however, you are advocating interference. You are for creating credit (essentially money out of thin air). Theft by credit expansion. You are against sound money.

    End the Fed. Honest money. No more belief in the church of Selgin the self-promoting interfering technocrat and his church's alleged "stable spending catechism."

    • George Selgin

      jandr0, with all due respect, were you to read Hayek's earlier (1920s and 1930s) works on monetary economics, and not just his "Fatal Conceit," you would know that he was no less an advocate of a stable spending norm than I am. So were plenty of other well-known monetary economists, so your suggestion that I imagine myself to be speaking from on-high is silly.

      Yet though I am not original, my arguments are at least arguments, and not mere empty slogans, such as "let's have honest money," "end the Fed," or "don't let banks create credit "out of thin air." Or rather, they suggest that a constant credit or money stock is the cat's meow, which is to say, something that generally just ain't so. Ultimately such naive opinions do more harm than good to the cause of market-oriented monetary reform.

      Consider your statement that firms "do not deserve to be in business" even when what forces them out is a collapse of spending. So in your view it was the fault of businessmen–all of the!–that so many failed in the 1930s, or in 2008-9 — both occasions when the Fed allowed spending to shrink? With enemies like you, the Fed needs no friends!

  • Spencer Hall

    You don't tell the world to "maintain a stable flow of spending", esp. considering secular strangulation. Like I said, then inflation will consume us: as lending by the DFIs is inflationary, but lending by the NBFIs is non-inflationary. If you don't understand economics, you shouldn't have a blog.

    As the "Bank Credit Analyst" Research wrote: It takes increasing infusions of Reserve Bank credit to generate the same inflation adjusted dollar amounts of GNP.

    • George Selgin

      You miss the point entirely about inflation and stagnation: if indeed goods' unit costs of production are rising, there is no harm, and even considerable merit in, a policy that allows their prices to rise to reflect that fact. If goods' costs rise so much that this leads to severe inflation, the problem isn't stable spending: its the stagnation itself! Your "solution" of reducing spending to keep prices down would make life even more, not less, miserable. In short, you don't "fix" an adverse supply shock–a greater scarcity of goods–by making money scarcer as well.

      And if you can't be reasonably polite, you shouldn't comment on others' blogs. Nor will you comment on this one any longer!

      • Spencer Hall

        You have it wrong again. My solution doesn't reduce spending, it increases spending. Lending by commercial banks is inflationary. Lending by non-bank conduits is non-inflationary. True, with limited upward and downward price flexibility, that unless money flows expand at least at the rate prices are being pushed up, incomes will fall, output can't be sold, and jobs will be lost.

        But there's no assurance that an increase in new money will be matched in the market place with an offsetting addition to the supply of goods and services. Quite the opposite of all experience.

        • Spencer Hall

          I've hunted and fished most my life. You have to be able to observe to track. What's happened is obvious. FDIC insurance was increased from 40,000 to 100,000, then from 100,000 to 250,000. The proportion of time (savings) deposits has vastly increased. Every time the ratio increases, gDp falls.