Interesting discussion about the multiplicity of equilibria in financial markets, in particular in bond markets! I learned a lot from digging into the relevant papers. I think I spot a big difference between macroeconomists and financial economists. I used to belong to the first tribe, but maybe not anymore :) Financial economists price long-dated assets, like government bonds or corporate bonds, without worrying too much about which equilibrium bondholders are coordinating on. They basically just try to compute the present discounted value of the cash flows that have been promised as best they can. Bond investors who are pricing Microsoft's 40-year paper just project and discount the cash flows. They're not too worried about which equilibrium they're all going to coordinate on. Take any valuation or asset pricing textbook, and I think you'd be hard-pressed to find anything about the multiplicity of equilibria other than maybe a discussion of bank runs. Read @JohnHCochrane 's Asset Pricing cover to cover (Darrell's book was too hard for me): no equilibrium selection anywhere, at least I don't recall seeing it. Finance doesn't treat it as first-order for valuing long-dated claims. (Of course, when John wandered over into macro, he spent a lot of time thinking about equilibrium selection, but that is consistent with the point I'm trying to make. ) Same goes for John Campbell's Financial Decisions and Markets. And Campbell is not a Chicago economist. He built the excess-volatility literature. Even when finance concludes prices deviate from fundamentals, we reach for discount rates, sentiment, limits to arbitrage. Not equilibrium selection. The credit literature prices long-dated defaultable debt daily ( Merton, Duffie-Singleton, and Pan-Singleton on EM sovereign CDS ) with default intensities driven by fundamentals. No sunspots. When financial economists do focus on self-fulfilling dynamics, it's mostly on funding and liquidity at short horizons, as in Brunnermeier and Pedersen. Macroeconomists have a different tradition. The sovereign-debt literature has emphasized multiple equilibria going back to models like Cole-Kehoe (a great paper by 2 amazing economists, one of who is my long-time coauthor and mentor). That creates an important role for policymakers: eliminate the bad equilibrium and coordinate markets on the good one. (I secretly suspect that's why macroeconomists like this.) In the Eurozone, the example that people always go back to is Mario Draghi's famous 2012 "whatever it takes" speech. All he had to do is speak those words and the Eurozone ended up in a virtuous equilibrium, where sovereign spreads were much lower. No bonds were ever bought under the OMT program. Costless equilibrium selection. Not quite. Look at what actually happened over the following decade. The ECB ended up running a very large balance sheet and rolling out several programs with increasingly complicated acronyms, all of which helped sustain low sovereign funding costs in the supposedly virtuous equilibrium. In the process the ECB was engineering was large cross-country transfers, transfers that I have documented in my work with Yili Chien and Zhengyang Jiang and Matteo Leombroni. nber.org/papers/w34311 So actually, if you look at the Eurozone evidence closely, you realize that the ECB re-engineered the underlying cash flows pretty dramatically. So not just a matter of picking a virtuous equilibrium, but actually a matter of reallocating resources across countries in state contingent ways. Draghi's announcement was a state-contingent promise of transfers.** Sometimes it's enough to just follow the cash flows. Here's why I think this matters. The multiple equilibrium doctrine is part of the official line at the ECB which suppresses the price signals that would maybe force a country like France to actually implement serious fiscal reforms. It kills the only enforcement mechanism we have. If the ECB delivers low funding costs for all governments, that significantly reduces the probability of fiscal reform imo. As Olivier pointed out, counting on the ECB to intervene would be unwise, but I do think the ECB has set some unfortunate precedents in this regard. **Valentin Haddad, Alan Moreira, and Tyler Muir make this point about the Fed's 2020 corporate bond backstop ("Whatever It Takes? The Impact of Conditional Policy Promises"): prices jumped on announcement, purchases were trivial, and the response is what you get when the market prices a conditional promise; a put written on the bad states. You don't need equilibrium selection to explain announcement effects. aeaweb.org/articles?id=10.12…