The Neo-Fisherian Counterfeiter

Just a thought-experiment that's been running through my head.

You are a saver and lender. You take some of the money you have earned as income, that you don't want to spend yourself, and lend it to people who want to borrow so they can spend it instead of you. Suppose you want to lend more than people want to borrow at the current rate of interest you charge, and you are unwilling to lend to riskier borrowers. You decide to lower the rate of interest you charge, so people will want to borrow more from you. You face a downward-sloping demand curve for loans. If you want to save and lend more of your income, you will need to lower the rate of interest you charge your borrowers. The greater the substitutibility between you and other lenders, and the smaller you are in the total market for loans, the more elastic will be the demand curve you face. Only in the limit, as the market for loans approaches perfect competition, does your downward-sloping demand curve become perfectly elastic at the market rate of interest.

Maybe you start borrowing from other savers, so you lend even more. You are now a financial intermediary, because you both borrow and lend.

Now suppose you go over to the Dark Side. You become a counterfeiter. You print money that is indistinguishable from central bank money. You lend out not only the money you have saved from your income, and the money you have borrowed from other savers, but you also lend out the money you have printed. The central bank knows there's a counterfeiter out there somewhere, but doesn't know it's you. And the central bank wants to keep inflation at the 2% target, so adjusts its own money printing one-for-one to your money printing, so your counterfeiting activities have no effect on the money supply growth rate and no effect on the inflation rate. You still face a downward-sloping demand curve for loans. If you want to print more money and lend more money, you will need to lower the rate of interest you charge your borrowers.

Now suppose the central bank shuts down its own printing presses. The stock of genuine central bank money stays constant. The only new money that enters the economy is the money you print and lend out. At the margin, you are the central bank, but nobody knows it. The faster you print money and lend it out, the faster the total stock of money grows, and the higher the inflation rate, and the higher the expected inflation rate, and the higher are nominal interest rates across the whole economy. There is an upward-sloping relationship between the nominal rate of interest and the rate at which you print and lend money.

Does this upward-sloping relationship mean that if you want to print and lend more money you must now raise the nominal rate of interest you charge on loans? No. Because as far as everyone else is concerned, you are just another lender of money, not the ultimate printer of money. They see no link between anything you do and the rate of inflation. They see you being affected by the rate of inflation, just like everyone else. If you want to print and lend more money you must first lower the rate of interest you charge borrowers, so they want to borrow more. You still face a downward-sloping demand curve for loans.

Now suppose your counterfeiting gets discovered. Everyone knows you are the one that's printing money, and knows that the faster you print money the higher will be the inflation rate, other things equal. But you hire an excellent lawyer who discovers an obscure law that lets you, and only you, go on doing what you have been doing. (OK, but it's only a thought-experiment.)

Does anything change?

If people observe you printing and lending more money, they know that will cause inflation and nominal interest rates to be higher than they otherwise would be. But they do not know that it will cause inflation to be higher in future than it was in the past. Maybe the demand for money has been growing more quickly, along with the supply. Other things might not be equal.

But if you knew that other things were equal, and you simply wanted to print and lend more money because you wanted to cause higher inflation, so you wanted people to borrow more money from you, what would you do? Would you raise or lower the nominal rate of interest you charge on loans?

The simplest answer is "neither". You simply tell people you are going to be printing more money because you want higher inflation, start printing and lending more money to whichever (safe) borrowers bid the highest, and let the market figure out what interest rate you can get on your loans. Auction them off.

But that simple answer is not allowed. The only thing you are allowed to announce is the nominal rate of interest you charge for loans. Do you raise it or lower it? Do you face an upward- or downward-sloping demand curve for your loans?

It depends on how people interpret your action. What information do they learn when they see you raise the nominal interest rate? That depends on why they think you did it.

Suppose people see you raise the nominal interest rate you charge on loans. One plausible interpretation is that people think that you, or other savers, wanted to save less, or that borrowers wanted to borrow more, so you simply raised the rate of interest to match the new equilibrium so you didn't have to print more money. And if people do interpret your raising the interest rate that way, your action will have exactly the opposite effect than the one you wanted. Because people will save more, and borrow less, so you actually print less money, and so inflation falls, compared to what would otherwise have happened.

How plausible is the Neo-Fisherian interpretation of your action? How likely are people to think "He's raising the nominal interest rate he charges on loans, so he must be trying to tell us he wants to print and lend more money and raise the inflation rate, so we know that inflation will rise, so we will need to borrow more money from him to make our stocks of money grow more quickly in line with the higher inflation rate"? Especially if, with a negative interest-elasticity of the demand for money, an increase in the nominal interest rate will mean a smaller real stock of money demanded, so the immediate effect would be for people to pay back some portion of their existing loans.

Either way, setting a nominal interest rate is a silly way to conduct monetary policy. It's a rotten communications strategy for central banks.

Comments

[dlr]

What information do they learn when they see you raise the nominal interest rate? That depends on why they think you did it.

Even Chair of the neo-fisherite council John Cochrane is coming around to the view the interest rates can be very murky focal point on the way to equation equilibrium:

The big question is expectations. Will people read higher interest rates as a warning of inflation about to break out, or as a sign that inflation will be even lower?

http://johnhcochrane.blogspot.com/2016/03/neo-fisherian-caveats.html#more

[Majromax]

This line of thinking might lead to an empirical question: in a VAR model, do unexpected changes in the central bank rate or a typical private-sector rate (say the 5-year mortgage rate) have a stronger impact on M1/2, NGDP, and the price level?

Your core point here is that interest rates are an ambiguous marker of the rate of change of the money supply, but to get there you went over a secondary point that the money circulating in private hands depends on the actual loans on offer rather than a central bank benchmark rate. Once your CB shuts down its printing presses, it can set whatever benchmark rate it wants but it will be wholly ineffective at influencing the private-sector (that is, counterfeiter's) rate.

This is more than just a theoretical question, as bank prime rates do not have a 1:1 correspondence with central bank moves. The 2015 Bank of Candada rate cuts (0.5%) were met with only 0.3% in cuts to bank prime rates. Even if the long-run equilibrium is a competitive one, imperfect competition could cause available private-sector rates to change more slowly than BoC rates.

[Market Fiscalist]

I know this is cheating but I think it may be inline with what some neo-Fisherians think. Couldn't I raise interest rates and at the same time print up the money and spend it on goods and services (or just give it away) and quite easily get both higher rates and higher inflation?

[Mike Sproul]

This is the kind of trouble you get into when you think that paper money has value because of its supply and demand, and not because of the assets backing it. Counterfeiters issue money without backing it, hence they cause inflation.

[Too Much Fed]

Mike Sproul said: "Counterfeiters issue money without backing it, hence they cause inflation."

Can the counterfeiters cause price inflation even if their "money" is backed?

[Too Much Fed]

Nick, let's say the counterfeiters produce a different type of "money". The central bank and the counterfeiters both permanently fix their "money" to each other (what you would describe as both ways).

How does the system work now?

[jonathan]

"You still face a downward-sloping demand curve for loans."

Wait, why?

Suppose you make a one-dollar loan. This involves creating a dollar. If the central bank wants to keep the money supply constant, it reverses this by selling a one-dollar bond. These bonds are bought by some other saver/lender in the economy, instead of making a loan to some other borrower. Then there's one more borrower out there who wants to borrow from you.

Thus the act of printing money and making a loan (buying an asset) causes the Fed to issue an asset that exactly meets the demand for assets you just generated. Effectively the Fed has increased its borrowing. We can even cut out the middle man by assuming that your "loan" is actually just buying a bond from the Fed. Then you're just printing money and lending it to the Fed.

So the demand curve you face isn't downward sloping. Unless I'm confused.

[Ralph Musgrave]

Nick Rowe says his idea is a thought experiment. It’s much more than that: it’s a description of the real world, in the following sense.

Nick correctly says that given a counterfeiter, the central bank has to withdraw CB money from the private sector dollar for dollar. Quite. And in the real world we have non-government organisations that print money: they’re called “private banks”. That is, private banks do not get all the money they lend out from savers: they also print a fair amount of new money every year.

If private banks were to significantly increase the amount they print and lend, then governments would need to confiscate base money from taxpayers to compensate, or implement some other deflationary measure. I.e. taxpayers subsidise the sort of respectable private money printing that private banks engage in in much the same way as taxpayers subsidise backstreet counterfeiters.

George Selgin described the orgy of theft of taxpayers’ money by private banks that would take place where private fractional reserve banks are introduced to a “base money only” economy at the link just below. (You can miss out his first two paras and start at “Perhaps the simplest…”)

http://capitalismmagazine.com/2012/06/is-fractional-reserve-banking-inflationary/
Also I recently put a paper online entitled “Taxpayers subsidise private money creation”:

http://www.scribd.com/doc/306133614/PrivateMoney-ToMunichPDverson

[marcus nunes]

Nick, Apt post for "April Fool´s day"!

[Nick Rowe]

dlr: Thanks for that link. I had read John Cochrane's previous Neo-Fisherian post, but not that one. He is not so much "chair" of the neo-fisherite council, but one of the more moderate members, (on a good day). He is sorta getting at the same thing right at the end of his post, when he says "The big question is expectations. Will people read higher interest rates as a warning of inflation about to break out, or as a sign that inflation will be even lower?" That's a bit like my "It depends on how people interpret your action. What information do they learn when they see you raise the nominal interest rate? That depends on why they think you did it."

(His previous post ducks the real question, of what happens when the central bank *announces* that it will raise (future) nominal interest rates.)

Majro: I *think* that's right.

MF: then you have both monetary and fiscal policy happening at the same time, which complicates the question.

Mike: There's an old post by David Glasner (I did a followup) where he shows that even under the gold exchange standard central banks can influence the price level by changing the bank rate. The price of gold varies too, according to how much gold the world's central banks keep in reserve.

jonathan: you are not confused. If central bank and counterfeiter lend to the exact same type of borrower (same risk-class), and if the central bank withdraws money one-for-one with the counterfeiters printing money, then you get the limiting case where the countefeiter faces a perfectly elastic demand curve for loans. (OK, I was a little sloppy in presenting my thought-experiment, but the key point is that even in that special case the demand curve for the counterfeiter's loans does not slope up.)

Ralph (and TMF): Yep, it's much closer to the real world than my "here's a crazy thought-experiment" might imply. But there are two differences between my counterfeiter and a real-world commercial bank:

1. The money that commercial banks "print" (chequing account money) is not a perfect substitute for central bank currency. They are imperfect substitutes.

2. Even if commercial banks issued currency (like they used to) that is a perfect substitute for central bank currency, the two currencies have different names on them, and commercial banks are beta, who promise to redeem their money for alpha central bank money, not vice versa. My counterfeiter is alpha (or more precisely, a "sneaky fucker" (technical term) who makes no such commitment.

marcus: I probably shouldn't have posted it on April 1st! But I woke up too early, and hit "post".

[Tel]

I always understood the argument you are using as the reason why the Federal Reserve should be at least partly private and governance be driven by the banks themselves. The right way to think about the Federal Reserve system is as a cartel, rather than as a government department (although there is crossover because the Fed depends on government to maintain its currency monopoly).

Thus, the bankers will grow the currency as much as it is profitable to do so, until they reach the point where further growth would start eating into the value of their portfolios. Then government (and Keynesian economists) push them for a bit more than that so government spending is covered with cheap loans and that's about where it comes to an equilibrium.

If you look at the chart of the 30 year mortgage rates in the USA, they are slowly but surely heading to zero, but of course we all know it cannot be allowed that home loans will ever be given away at zero interest... so there will be a break in that trend.

FRED: 30-Year Conventional Mortgage Rate

[Mike Sproul]

Too much fed:
"Can the counterfeiters cause price inflation even if their "money" is backed?"

If the counterfeiter's money is backed, then he's not a counterfeiter.

If the fed issues $100 in FRN's, backed by assets worth 100 oz of silver, then $1=1 oz.
If the counterfeiter issues $60 perfect copies of FRN's, and backs them with 60 oz worth of assets, then the real $ and the fake $ are still worth 1 oz. each.

If the counterfeiter issues $60 perfect copies, and does not back them, then the $160 in circulation are backed by only 100 oz worth of assets, so $1=.625 oz.

[Philip George]

You are completely right in reiterating that the demand curve for loans is always downward sloping.

So why does the graph of Industrial and Commercial Loans vs the Effective Federal Funds Rate seem to point in the opposite direction? See https://research.stlouisfed.org/fred2/graph/fredgraph.png?g=3I8H

It clearly shows that whenever the Fed raises interest rates from close to zero levels the quantum of industrial and commercial loans goes up, not down.

The reason is that in your entire argument you have ignored the supply side of the equation.

Very low interest rates make it possible for players in the financial asset markets to earn high returns by using high leverage. When interest rates are raised leverage becomes expensive and returns fall drastically. So financial companies are forced to look at alternative places to deploy their money. This is why loans to the real side of the economy increase when rates are raised from very low levels.

I wrote about this five years ago at http://www.philipji.com/item/2011-07-24/why-banks-do-not-lend-at-near-zero-interest-rates

At that time I was still groping for the correct model of money which I have since written about in my book "Macroeconomics Redefined" http://www.amazon.com/dp/B00ZX9O5XQ

Incidentally, the book predicts a massive crash by the end of this year.

[Ralph Musgrave]

"Either way, setting a nominal interest rate is a silly way to conduct monetary policy. It's a rotten communications strategy for central banks." In addition to which, there is no obvious reason why, given a recession, borrowing, lending and investment should be increased, any more than expenditure on education, tomato sauce, restaurant meals or health care should be increased.

[Nick Rowe]

Philip: All I really need is that the excess demand curve (quantity demanded minus quantity supplied) for loans slopes down. So if my counterfeiter wants to print and lend more money, he lowers the rate of interest he charges borrowers and pays lenders. So you are right I ignored the supply side, but it was simpler my way.

Ralph: There is an obvious (to me) reason why, in a recession, the supply of money should be increased. Whether that results in increased expenditure on tomato sauce or on something else doesn't concern me. (Though tomato sauce (being a Brit you mean "ketchup") is currently a political thing in Canada right now, for reasons I don't quite understand.)

[Too Much Fed]

Mike S. said: "If the counterfeiter's money is backed, then he's not a counterfeiter.

If the fed issues $100 in FRN's, backed by assets worth 100 oz of silver, then $1=1 oz.
If the counterfeiter issues $60 perfect copies of FRN's, and backs them with 60 oz worth of assets, then the real $ and the fake $ are still worth 1 oz. each."

Let's permanently fix silver and currency at $1 = 1 oz "both ways". Let's make silver and currency both MOA and MOE. Hold quantities of goods and services fixed. Something changes so silver can be mined with no cost. Silver increases by 10%. People take this silver to the counterfeiter who makes perfect copies of currency. Since goods/services quantities are fixed, price inflation rises.

The price inflation would be both in terms of silver *and* currency.

[Too Much Fed]

Nick said: "My counterfeiter is alpha who makes no such commitment."

Let's have the central bank and counterfeiter both agree to a fixed exchange rate both ways (the counterfeiter also makes the commitment). It just happens to be 1 to 1.

Are they both alpha's now?

[Gizzard]

You are not describing how banking really works if you are describing it as a borrower getting money from a saver. No one taking out a loan keeps me from using my savings. If I were actually lending out my savings then I wouldn't be able to spend when someone borrowed it, I would have to wait for it to be paid back.

When are you going to start understanding what banks do?

[john]

In our current below capacity predicament, you can offer your money at almost any low rate, but without expected ROI, borrowers may not borrow, other than financial arbitrage.

You would do better to print and spend directly.

john

[Jacques René Giguère]

Thanks Nick for giving me the only topic on this thread where I can meaningfully contribute but please do not confuse ketchup and tomato sauce. You can't put sauce on a hamburger nor make spaghetti sauce with ketchup...

[Scott Sumner]

Excellent post, especially the concluding sentence.

[Tel]

Phillip:

It clearly shows that whenever the Fed raises interest rates from close to zero levels the quantum of industrial and commercial loans goes up, not down.

Looks to me like change in the volume of loans is the leading indicator, and interest rates are lagging behind by approx 6 months. Try the same data averaged down to quarterly samples and you see that most of the peaks and the bottoms show the blue line turns around ahead of the red line. The exceptional situation was 2008, and of course around 2010 we expect the red line to start coming back up again but it never did (Bernanke's great Keynesian experiment, which IMHO has resoundingly failed).

I would argue that the Fed deliberately adapts the supply side to fit what they perceive to be changes in demand (but they are a bit slow to react every time). The banking industry operating as a whole (including the counterfeiter) makes profit [A] from printing money and [B] from interest on loans... so you can see there's an optimal balance because if you print too much money too quickly your loan portfolio becomes worthless.

I added CPI to the graph, because (at least in theory) too much counterfeiting should start to show up in price inflation, although we can then get into Cantillon effects, who gets first dibs on the new money (typically whoever has best political connections) whether CPI is a quality measure of prices and a bunch of other stuff. At any rate, the price-inflation part of the equation cannot entirely be ignored. PS: also adjusted the graph colours to give more of a "tomato feel" to it.

[Nick Rowe]

TMF: with a bilateral peg, they are both alphas or both betas. It's an unstable arrangement.

Gizzard: I am not following you. Consolidate the commercial banks plus central bank. The consolidated banking system: borrows money, lends money, and creates (also destroys) money. Just like my counterfeiter.

john: whether or not borrowers want to borrow, and at what rate, depends on expected future monetary policy.

Jacques Rene: the Brits use language differently. What we call "Tomato ketchup" they (usually) call "tomato sauce", and they put it on their chips (fries). I can't remember what they call tomato sauce, that they use to make spaghetti sauce.

Scott: thanks!

Tel: "I would argue that the Fed deliberately adapts the supply side to fit what they perceive to be changes in demand (but they are a bit slow to react every time)."

That sounds plausible to me.

[Mike Sproul]

Too much fed:
With 10% more silver, the value of silver would fall by something around 10%, and since the dollar is pegged to the ounce, the dollar loses 10% of its value too.

But then if the counterfeiter issues more currency without backing it, the dollar would fall relative to silver. This would happen because the dollar has less backing, not because there are more dollars relative to goods.

[Too Much Fed]

Nick said: "TMF: with a bilateral peg, they are both alphas or both betas. It's an unstable arrangement."

I'd say they are both alphas. Why is it an unstable arrangement?

[Too Much Fed]

Mike S said: "With 10% more silver, the value of silver would fall by something around 10%, and since the dollar is pegged to the ounce, the dollar loses 10% of its value too."

Sounds good. Since I made silver and currency both MOA and MOE, silver and currency do not fall in value. The goods and services "rise in value" (price inflation).

So it looks to me that the answer to my question is yes, price inflation can happen even though the dollars are fully backed.

And, "But then if the counterfeiter issues more currency without backing it, the dollar would fall relative to silver. This would happen because the dollar has less backing, not because there are more dollars relative to goods."

It looks to me like it is both. More dollars with no fixed rate to silver means the dollar falls relative to silver. More dollars with no fixed rate to goods/services means the dollar falls relative to goods/services (price inflation).

[Too Much Fed]

Nick said: "You simply tell people you are going to be printing more money because you want higher inflation, start printing and lending more money to whichever (safe) borrowers bid the highest"

That means *new* "money" for *new* bonds/loans, right?

[Jacques René Giguère]

Nick: You're right. My old memory is now clicking back to the time I was a student in Blighty. Seems that in self-preservation my brain had blocked all souvenirs of British universities cafeterias.
Back to monetary theory.