Think back to philosophy class:
Would you ever believe the promise of an Act-Utilitarian? No. An Act-Utilitarian will perform action A if and only if action A will maximise the sum of expected present and future utilities. His past act of having promised to do A is not a reason for the Act-Utilitarian to perform action A in future. He will perform or not perform action A regardless of whether in the past he promised or did not promise to perform action A. (Time consistency, rules vs discretion, and all that.)
The above is an argument against Act-Utilitarianism as a theory of ethics. The argument seeks to show that Act-Utilitarianism is inconsistent with the institution of promising.
My argument against the Fiscal Theory of the Price Level is similar. Because a bond is a promise. But it cuts deeper. Because an Act-Utilitarian could try to defend Act-Utilitarianism by arguing that the institution of promising is not essential to ethics. An FTPLer could not defend FTPL by arguing that bonds are not essential to FTPL. With no bonds, there is no FTPL. Because FTPL is about bonds, and about what determines the (real) value of those bonds.
The whole point about promises is that future actions are not exogenous with respect to past promises. Bonds are promises. The whole point about bonds is that future payments to bondholders are not exogenous with respect to past loans. FTPL says they are exogenous.
I give you $100. Next year you give me $105. Does that mean I lent you $100 and you repaid the loan by giving me $105? Not necessarily. Maybe you are my friend, and I simply gave you $100 because I saw you needed it a lot more than me. And next year you gave me $105 because you saw I needed it a lot more than you. What makes it a loan is if the two actions are linked. I gave you $100 because you promised to give me $105 next year if and only if I gave you $100 this year. You give me $105 next year because I gave you $100 this year. Your giving me $105 next year is not exogenous with respect to my giving you $100 this year. That's why I gave you the $100 in the first place. Because I thought it wasn't exogenous. If I had thought it was exogenous, I would never have given you the $100. Unless it were a gift.
FTPL says that B(t)/P(t) = EPV[S(t)], and that EPV[S(t)] is exogenous with respect to B(t)/P(t). B(t)/P(t) is the current value of the loan, and EPV[S(t)] is the expected present value of the repayments. The S(t+1)=$105 that you will give me in future is exogenous with respect to the B(t)=$100 that I gave you in the past. But if that were true, how did B(t) ever get to be positive in the first place? What idiot ever gave this government the first loan? It can't have been a loan; it must have been a gift.
Start out at t=0 and assume B(0)=0 and S(t)=0 for all t. The government has no bonds, never borrows or lends, and has neither primary surpluses or deficits. Assume it is common knowledge that FTPL is true for this government. How could B(t)/P(t) ever become strictly positive? How could FTPL ever get started?
Suppose there were some exogenous shock to S(t), for some future t. So that S(t) became strictly positive. "Hey everyone!" the government says, "I'm going to be giving away $105 next year! I just can't help myself! I've got an uncontrollable urge to give $105 to whoever is holding this bit of paper next year. Anyone want to buy it? Sold for $100 to the highest bidder!" And what would the government do with that $100? It would spend it of course.
So an exogenous shock that makes S(1) increase to $105 causes B(0)/P(0) to rise to $100, and this in turn causes S(0) to fall to -$100.
The whole story is daft. What exogenous shock would cause a government to want to give away $105 next year? And why would it want to give $105 to the person who happened to be clutching a bit of paper called a "bond"? Why not give it to someone else instead, like the government's favourite charity?
Suppose that half the bonds were destroyed in a fire, and nobody could remember who owned them. Would the government double the real payments to the remaining bondholders? That's what FTPL says. That's what happens in a business that goes bankrupt. But if people had expected the business to go bankrupt, they wouldn't have bought the bonds in the first place. FTPL is a special theory about the bonds of governments that unexpectedly go bankrupt.
An Act-Utilitarian would never do what he promised to do simply because he promised to do it, unless he wanted to fool people into thinking he was not an Act-Utilitarian. A loan is a promise. An FTPL government would never repay a loan simply because it was a loan, unless it wanted to fool people into thinking it was not an FTPL government. But FTPL assumes that people know the government is in fact FTPL. It's supposed to be a rational expectations equilibrium. That's the whole point of FTPL. It's people expectations of those exogenous primary surpluses that determine the real value of the current stock of bonds. But if those expectations were exogenous, why would an FTPL government ever choose to run a primary surplus and give anything to bondholders? And why would anyone buy a bond expecting it to do this?
If it were common knowledge that FTPL were always and everywhere true, FTPL would be irrelevant, because the stock of government bonds would be always and everywhere zero. FTPL could never get started.
We can imagine a government that starts out as non-FTPL, or believed to be non-FTPL, and which borrows. B(t)/P(t) is now positive. But then something changes, and the government becomes FTPL. Or people are surprised to learn that the government was in fact FTPL all along. But if people did learn, with certainty, that the government was in fact FTPL, B(t)/P(t) would immediately fall to zero. Because there is no reason whatsoever that a government known with certainty to be FTPL would ever give away anything valuable to people clutching little bits of paper, simply because they were clutching those little bits of paper.
FTPL is a parasitic theory. It is a theory that cannot be universally true. It needs a host, and must remain hidden. Just like the secret Act-Utilitarian who makes promises that have no effect on his future actions. He makes promises only to decieve, and he only keeps them if he would have performed the promised action anyway, or to prevent people from learning that he is an Act-Utilitarian.
Worthwhile Canadian Initiative 
Miami: “Have you even bothered to ask yourself questions like what does the ftpl say about what happens if you double the growth rate of M or changes the inflation target?”
Yes. And answered it. See my previous post.
And you clearly have misunderstood my post.
Min: “Reasoning back from a future date uncertain is problematic.”
Yep.
Jussi: “If yes, what limits here the real resources the government can use?”
Yes, of course. Increased government spending is not inflationary, unless the central bank makes it so. The limit is the government’s ability to collect tax revenues to pay the interest on what it has borrowed.
@Nick
Sure, I get that part. But in my fiction the Central Bank issues debt (as interest paying reserves), pays the interest in order to monetize government spending. Here the amount of government spending cannot be limited by tax revenues as the government is fully monetized by the Central Bank and the inflation front is taken care by the interest on reserves.
What do I miss – what is the limit then?
If you are printing money to pay interest on money, the growth rate of the money supply equals the interest rate, and equals the inflation rate. Because the stock of money is growing over time. See my old post “when new money is paid as interest on old money”.
@Nick
Yes, that was a good teaching post. I reread it.
But would it mean by the same token the debt issued by the Treasury is inflationary if it was taken to pay the interest? But that doesn’t sound true. But then again why would it matter which entity issued the debt (interest paying reserves or Treasury debt)?
Bullshit. Monetary policy confined by existing accounting and institutional arrangements gives you the ftpl. Saying its not “true” doesn’t even make sense if you believe the QTM to be true. They map directly onto each other if you remain confined to reality.
“What was really new about the FTPL compared to standard theory is that it insists that the government budget constraint, in economies where large amounts of nominal interest-bearing debt are issued, has to be treated as one of the central equations determining equilibrium and determining the price level. Once you realize that, you realize that there is a kind of a simple connection between outstanding nominal interest-bearing liabilities, taxes and the price level that’s comparable to the simple connection between outstanding money balances, demand for money and the price level.
These relationships are both true. That is, it is true that in equilibrium, the total nominal value of the debt, the total real value of interest-bearing debt, has to match expectations of future tax backing for that debt. And that’s true in any model.
And then at the same time, it’s true, of course, that in equilibrium the real balances held by the public have to match the demand for those real balances. It can be helpful, depending on the circumstances and as a crude first approximation, to think of the price level as being determined by the ratio of the nominal quantity of money to the real demand for money. Or it can be useful to think of the price level as being the ratio of the nominal value of interest-bearing debt outstanding to the tax backing for that debt. It’s hard for people who get used to thinking about it in one simple way or another to realize that in equilibrium both these ways of thinking about it are correct.”
-Chris Sims
You say that’s wrong but have yet to demonstrate how its wrong. Just that you don’t want to live in a world like that. Whatever that means.
“So an exogenous shock that makes S(1) increase to $105 causes B(0)/P(0) to rise to $100, and this in turn causes S(0) to fall to -$100.”
Nope. S(0) is about expectations, B(0) is representative of the -$100.
“So an exogenous shock that makes S(1) increase to $105 causes B(0)/P(0) to rise to $100, and this in turn causes S(0) to fall to -$100.”
No, S(0) is $105. Its about expectations. The -$100 is represented by B(0).
Sorry for the double post.
Jussi: if you issue new debt to pay interest on old debt, and keep on doing it forever, you are running a ponzi scheme. Is that ponzi scheme sustainable? If it’s not sustainable, it must lead either to default, or else you raise taxes and stop doing it. With currency it is sustainable, because people will hold currency even if the real rate of return on currency is very negative (there’s inflation). Plus, we measure prices in terms of currency, not in bonds. (Currency is the unit of account.) So if people expect default on bonds, bond prices drop relative to currency, but the central bank does not need to let the price of currency drop relative to goods (inflation).
Miami: “You say that’s wrong but have yet to demonstrate how its wrong. Just that you don’t want to live in a world like that. Whatever that means.”
I live in a world (called “Canada”) where a central bank adjusts the supply of money to the demand for money to ensure inflation remains at 2%, and where the federal government adjusts expected future fiscal policy to match the total stock of bonds outstanding, to prevent default, given that policy of the Bank of Canada. I can imagine a different world (Zimbabwe), where the central bank adjusts the supply of money, and inflation, to prevent default on bonds, taking expected future fiscal policy as given. But I don’t live in a world like that, nor do I want to.
Miami: let me try it this way.
Chris Sims, in that passage you quote, forgets that bonds can default, and that (fiat) currency cannot default. Currency is alpha, bonds are beta, because bonds are a promise to pay currency, not vice versa. And we define “inflation” as a falling price of currency against goods, not a falling price of bonds against goods.
The central bank that issues currency can, if it wishes, adjust the supply of currency to the demand for currency, to prevent the price of currency falling against goods. If it does that, the issuer of bonds has a choice: either adjust future taxes to prevent risk of default; or else allow the risk of default to increase, and let the price of bonds fall against currency, and against goods.
It is only if the central bank adjusts the supply of currency to prevent default on bonds, that expected future taxes influence the price of currency against goods.
“Inflation” means “the price of currency is falling against goods”.
If an issuer of currency promises (say) 2% inflation, then “currency default” means “inflation above 2%”.
An issuer of bonds promises to pay currency. A “bond default” means the price of bonds falls against currency.
Who prevents a bond default? Is it the issuer of bonds, or the issuer of currency?
If it is the issuer of currency, then the issuer of currency may need to let currency default (let inflation rise above 2%) to prevent a bond default.
But in sensible countries, with inflation targeting (or similar) central banks, it is the issuer of bonds who prevents bond default, by adjusting expected future taxes and spending to ensure it doesn’t happen.
@Nick
Thanks once again – being patience and all.
I have to still keep commenting as I would really like to understand why I seem to fail (am I the only one?) to get the money/debt relationship. I guess it is a lot about seeing how indifferent banks are between reserves and short term government debt, even in the event of liquidity crisis. And knowing the reserves can only be held by banks and government entities (or the likes) anyway.
I’m not sure whether you were rhetoric or not? As I see it both currency and debt are liabilities of the government in the broad sense. I guess it is hard to default the currency without bonds but I think it is possible. Of course the governments usually default bonds only. But both can be defaulted in practice and it is not sensible to default either, so not sure it is a great theoretical argument (or is it, why?). Money certainly carries liquidity premium usually (but look Germany today). But these smell to me more like difference in scale than in kind? Plus piling debt over debt seems to often bolster confidence in bonds (all major economies in the current on going crisis).
I guess my question is whether Treasury debt is sustainable if new debt is issued to pay interest on old debt? Does this depend on any qualifications like if the banks are forced to use only Treasury notes (and so no other reserves) to clear daily imbalances? Or debt issued is only a promise to return reserves tomorrow, like 1-day Treasury bill?
You could say the central bank adjusts the number of bonds outstanding or money its the same coin just different sides.
I’m not sold on your alpha/beta distinction, especially given the fact that the rate on bonds was lower in the US than that on reserves for a period of the crisis maybe even still at times. If reserves were alpha that shouldn’t happen, should it? Money is a beta saving vehicle. Its a stop gap not an asset people hold for any other reason than liquidity.
I think you’ve stretched the definition of default too far. Rising bond yields is not the same thing as defaulting and you know that. Currency default is inflation above target? What do you call it when its below target?
I know you see a MASSIVE distinction between the ftpl and the QTM but I don’t and I’ve read all your posts and still don’t see the distinction you’re trying to articulate. Most likely that is due to a poverty in my understanding of both theories but if it were obvious I think you would have been able to show that by now. As someone from the concrete steppes it was refreshing to see expectations and interest rates as a poor indicator of the stance of monetary policy, represented mathematically. If it looks like a duck.
Thanks for your time and patience, I’m a big fan of your blog posts.