With France in crisis, and Germany in much worse economic & political situation than when OMT, PSPP or PEPP were announced, there is nobody left that can provide the necessary strong fiscal backing for the euro and the ECB in case of large losses or high inflation. @lugaricano

🛑🇨🇵 France is junk (BB-) at 166 bps

So why do I believe fiscal backing is key to monetary stability, when MMT believes it is not? Theory first, starting from the fundamentals. According to MMT, the state's paper money is worth something because taxes are payable in it, and the state accepts nothing else in payment. This idea is old and goes by the name of "chartalism"; it runs from Adam Smith to Knapp. What is interesting is that, when you read MMT, a large part of it is shared by the neoclassical counterpart of chartalism, the fiscal theory of the price level (FTPL). And yet the two reach almost exactly opposite conclusions. While the FTPL argues that the price level is ultimately pinned down by the government's fiscal position (and inflation by news about future deficits), MMT argues that the fiscal position itself never matters. What moves prices, for them, always operates in the present: a change in aggregate demand, a wage bargaining round, or, in Mosler's version, the prices the government itself agrees to pay when it spends. This may come as a surprise: how come a theory founded on the notion that taxes give value to money also argues that expectations about the state's future capacity to tax play no role in the price level? The fundamental reason comes down to methodology. Since Bob Lucas, neoclassical economists have embraced the idea that we need to be serious about incorporating agents' expectations of the future into our economic models. This makes neoclassical models hard to write and to read, because everything now requires solving a dynamic system with Bellman equations. For heterodox economists, this is the wrong road: in their view, rational expectations is a wrong description of people, and its formal apparatus crowds out reasoning about how the monetary system actually operates. For instance, here is, in their eyes, an important realization that the whole neoclassical profession has missed: since we left the gold standard for floating exchange rates, we have removed the constraint that used to require the state to come up with pieces of metal to meet redemption and preserve the stability of the currency. Tymoigne and Wray say it in one line: "For a sovereign nation that does not promise to peg, there is no process that can lead to involuntary default." Of course, there is a bit of truth to this. In the language of my first tweet: we have removed the liquidity constraint on central banks, and this matters. A floating currency regime indeed prevents a first-come-first-served run on the currency, with its forced timing where reserves hit zero on a date (Krugman 1979). So far, we agree. But there were really two runs stacked on top of each other. The first was on the gold itself: get to the window before the vault is empty. The second was on the value of the currency: selling it because you doubt what it will buy tomorrow. That run needs no window and is not abolished by floating the currency. MMT is methodologically blind to exactly this: removing the liquidity constraint does not remove the solvency constraint. The reason is subtle. In a dynamic model, there are two important forward-looking equations. First, through households' decisions about how much money to hold, inflation is itself directly forward-looking: the higher I expect inflation to be tomorrow, the less money I want to hold today, and the higher inflation is today. Second, by iterating the government budget constraint forward, the valuation of government debt (including money) is also forward-looking: I only accept a price today that is aligned with the real resources I expect to receive from holding it. Through the first equation, inflation can take an infinity of paths, many of which are hyperinflationary. In New Keynesian models, where the same forward-lookingness runs through interest rates rather than money holdings, the hyperinflationary paths are handled by the combination of two elements: (i) an over-reactive central bank, raising rates more than one-for-one with inflation, which turns any departure from target into an explosive path, and (ii) the assumption that the government passively adjusts its surpluses to keep its intertemporal budget constraint satisfied. Strictly speaking, the explosive paths are then not so much ruled out as dismissed: the model assumes inflation never embarks on one. This works, but notice the two assumptions doing the work: fiscal capacity is maintained by assumption, and the explosions are excluded by assumption. An alternative (and simpler) way out is to assume that the government is not passive: it fixes its path of taxes and spending in real terms. The government debt valuation equation then pins down the price level: it settles at whatever level is consistent with expectations of future deficits. If the government plans to spend more with no intention of ever raising the corresponding taxes to pay me back in real terms, the real value of money (and of nominal debt) has to decline today, through a higher price level. This is the fiscal theory of the price level. A third alternative, the one I find most realistic, is the "escape clause". The central bank's primary tool for controlling inflation is the short-term rate, but it also stands ready to buy back the money supply if inflation ever reaches a level it deems unacceptable (say 15%). By conditionally falling back on quantity management when needed, the central bank can itself rule out the hyperinflationary paths. The nice thing is that this is an out-of-equilibrium threat: because investors understand that the central bank can do it, inflation stays anchored and the threat never has to be executed. This is also why it is crucial for a central bank to remain economically solvent. Once the value of its assets (including the present value of future seigniorage revenues) falls below the value of the money outstanding, the escape clause is no longer credible, and hyperinflationary paths become feasible again unless the fiscal authority steps in to control deficits. And this is where deficits enter the picture. Every deficit adds to the stock of claims that the backstop may one day have to make good, in real terms. Covering a bigger stock is not just more of the same: raising taxes gets harder the higher taxes already are, economically and politically. So fiscal capacity is a resource, and deficits use it up quietly. Nothing needs to break today. Investors simply watch the distance between the promises outstanding and what the state could still credibly raise, and they price it. That is how deficits matter continuously: not by causing inflation this year, but by consuming the capacity that keeps every future year anchored. MMT, for its part, does propose an anchor of its own: the Job Guarantee. The state offers a job at a fixed wage to anyone who wants one, and that posted wage anchors the price system, much as the gold parity once did with a buffer stock of metal. Note that this anchor is true to their method: a posted price, operating entirely in the present, no expectations involved. Whether it would hold is a debate for another day. Today, no country operates one; so whatever currently anchors the dollar or the euro, and whatever justifies "not worrying about deficits", it is not the Job Guarantee. So, after spending quite a bit of time thinking about this question, I have concluded that many MMT insights are right (the role of taxes, the removal of the liquidity constraint), but that the theory is fundamentally missing the forward-looking dynamics needed to think about hyperinflation and about valuation. When they do think about hyperinflation, they treat an exogenous collapse of the state's ability to tax as its source, but they do not take the point to its logical conclusion: accumulated deficits must eventually be covered in real terms for price stability to hold, so each deficit eats into the very tax capacity whose collapse they fear, bringing the inflationary dynamics closer. One last remark, on why this debate never seems to resolve. As long as the backstop is credible, it is invisible, and the two frameworks predict the same data. They separate only when fiscal capacity is tested. I will come back to this in my third tweet. A final word on credit: none of the theory above is mine (although errors might be). It runs from Cagan on money demand to Leeper, Sims and Cochrane on fiscal backing, and Obstfeld and Rogoff (1983) on the escape clause. The best single guide to all of it, escape clause included, is Castillo-Martinez and Reis, "How Do Central Banks Control Inflation? A Guide for the Perplexed" (Journal of Economic Literature, 2026). (2/3)