The Royal Swedish Academy of Sciences gave this year's economics prize to Robert Hall of Stanford and Michael Woodford of Columbia "for microeconomic foundations of macroeconomic stabilisation."1 Naturally, the citation is drier than the achievement. Postwar macroeconomics ran on statistical relationships between aggregate indicators, but those relationships broke in the 1970s when inflation and unemployment rose together and the models said they should not. Hall and Woodford rebuilt the subject from its foundations, deriving the behavior of the whole economy from the decisions of the households and firms inside it. The committee's own summary of what that changed is that "households and businesses are forward looking when making decisions about savings or product prices."2 Once that is true, what a central bank is expected to do next becomes part of what its current policy does. The committee's claim for the consequence is that the two of them "developed new macroeconomic models that have significantly improved our ability to implement effective stabilisation policy."2
Only Surprises Move Spending
Hall's contribution rests on one paper, published in December 1978, and the committee cites it 24 times, more often than any other work of his.3 Before it, a forecaster modeled consumption as a function of current income and treated the relationship as stable enough to steer by. Hall asked a different question: what must that relationship look like if households are optimizing over a whole lifetime?
The answer was austere. If a household is smoothing consumption across its life, it has already built everything it expects into what it spends today. Nothing predictable is left. "No variable apart from current consumption should be of any value in predicting future consumption," Hall wrote.4 Changes in spending are therefore unforecastable; only an event nobody saw coming can move them.
The result is cited loosely, and Hall warned against the loose version in the paper itself. It is tempting to summarize the theory by saying that consumption is a random walk, he noted, and "this is not accurate."5 The theorem governs marginal utility. Consumption inherits the property only under added assumptions about how households value an extra dollar.
He then did the thing that distinguishes the paper from a clever derivation. He tested it. With Frederic Mishkin, he took the prediction to a panel of individual households, where about 80 percent behaved as the theory says and the rest "lived hand-to-mouth," spending what arrived when it arrived.6 Hall's split has organized a whole literature since, showing the hand-to-mouth minority is where fiscal transfers do their work.
Hall spent the following decade supplying pieces the modern framework would need. He put the intertemporal elasticity of substitution, which governs how much households shift consumption across time when interest rates move, so low that he would say only that it "is unlikely to be much above 0.1, and may well be zero."7 The committee's own summary rounds this to an estimate of "about 0.1," which is looser than what Hall actually claimed; he reported a ceiling he could not distinguish from nothing. He also measured price against marginal cost across American industry and found markups running from a little under 2 to a little under 4, which is evidence that firms set prices rather than accept them.8 That finding anticipated what came next. A firm with a markup has room to leave its price alone for a while, and a model in which prices move sluggishly because firms choose not to move them is a model in which monetary policy has real effects. Hall supplied the evidence that firms have that room.
From 1978, the year of the consumption paper, until 2024, Hall chaired the National Bureau of Economic Research's Business Cycle Dating Committee, the body that decides when American recessions begin and end.9 He proved that downturns arrive unannounced then spent 46 years dating them after the fact.
Expectations the Bank Chooses
Woodford is, in the committee's words, "the key architect behind the so-called New Keynesian model."10 The construction took most of a decade. Working alone and with Julio Rotemberg across a run of papers from 1992 to 1999, which the committee calls the "key steps forward," he assembled a model of the whole economy in which every equation descends from somebody's optimization problem.11 His 1997 paper built an estimated model derived from household and firm optimization rather than from fitted correlations, and judged policy by a welfare criterion drawn from household utility instead of by forecast error.12 A central bank with such a model ranks policy rules by how well off they leave households.
Economic Engines
The engine that emerged has three moving parts, and the first of them is Hall's. Household optimization gives a relation between spending today and the expected real interest rate, so demand depends on the whole expected path of rates rather than on today's. Firms that hold a markup and reset prices only occasionally give a relation between inflation and expected future inflation, which is why inflation today depends on what firms think is coming. A policy rule closes the system. Because two of the three equations contain expectations of the future, the central bank's anticipated conduct enters the economy's present, and the model cannot be solved without taking a view on what the public believes the bank will do.
Interest and Prices, published in 2003 and running to 808 pages, is where the framework became a working manual. It derives optimal interest-rate policy from "intertemporal equilibrium modeling, taking full account of the endogeneity of private-sector expectations."13 Expectations respond to the bank, so the bank is partly choosing them.
Two results from the book have shaped practice. The first revives a nineteenth-century idea: the benchmark for whether policy is tight or loose is what Wicksell called "the natural rate of interest, the equilibrium real rate of interest in the case of flexible wages and prices."14 That rate moves with productivity, demographics, and saving, so a central bank that holds its nominal rate still while the natural rate falls has tightened without acting. Every modern argument about r-star descends from this.
The second result is the one that reorganized central banking: "The effectiveness of monetary policy depends as much on the public's expectations about future policy as upon the bank's actual actions."15 It follows that a bank able to bind itself to a future course beats one that re-optimizes freely at each meeting since the promise itself moves behavior today. Discretion, which sounds like an advantage, is a handicap.
Woodford pushed that logic to its hardest case with Gauti Eggertsson in 2003, when short rates in Japan had reached zero and conventional policy had run out. They showed that buying assets accomplishes nothing by itself: open-market operations, unconventional ones included, "should be largely ineffective to the extent that they fail to change expectations regarding future policy."16 What works at zero is a promise about later. The damage from the lower bound, they showed, could be undone almost entirely with no new instrument, so long as the bank could bind itself credibly to a course that depends on where it has already been.17 A bank that pledges to keep rates low past the point its own rules would suggest raises expected inflation now, which lowers the real rate now (the only lever it has left). They were careful about the limit: "it is not optimal to commit to so much future inflation that the zero bound ceases to bind."
By 2012, from the Jackson Hole platform, Woodford had turned this into instructions. "The most effective form of forward guidance involves advance commitment to definite criteria for future policy decisions," he wrote, and the notion that balance-sheet policy can substitute for such a commitment "is mistaken."18 The distinction he was pressing had acquired a name that year, when Jeffrey Campbell, Charles Evans, Jonas Fisher, and Alejandro Justiniano separated Delphic guidance, which "publicly states a forecast of macroeconomic performance and likely or intended monetary policy actions," from Odyssean guidance, which "does publicly commit the policymaker, just as Odysseus committed himself to staying on his ship."19 A forecast and a promise are different instruments, and only the second ties anyone's hands.
Where the Framework Strains
Neither laureate has oversold the machinery, and the committee is candid about where it strains. Its scientific background lists the framework's open problems and gives one of them a name: in the standard model "the effects of announcements far into the future had unreasonably strong effects on demand and prices today."20 This is the forward-guidance puzzle. The model says a central bank can move today's economy enormously by talking about a date years away, and no central bank has ever found that lever so powerful. Marco Del Negro, Marc Giannoni, and Christina Patterson measured the gap against survey expectations and found that accommodative guidance moved output and inflation expectations by amounts "28 and 8 times smaller, respectively, than implied by standard models."21 Alisdair McKay, Emi Nakamura, and Jón Steinsson then showed the model repairs itself part of the way: allow for borrowing constraints and incomplete markets, which stop households acting on distant news, and the modeled power falls to roughly 40 percent of the textbook effect at five years and nearly nothing at ten.22 The committee closes each of these entries by noting that no single explanation has become a consensus.
Woodford himself went after the foundation. In 2013, he argued that the rational-expectations assumption underpinning his own framework "may seem unjustifiably strong, and under some circumstances, is a heroic assumption indeed," and showed that one can impose quite stringent rationality on beliefs without being able to conclude that the model's predictions follow.23 He has since built models in which people plan over finite horizons, borrowing their structure, as he puts it, from "the architecture of artificial intelligence programs for problems such as chess or go."24
What Central Banks Borrowed
Central banks now publish inflation targets, explain decisions against a stated objective, and set the expected path of policy as an instrument. The Academy endorses that inheritance in its own voice, citing Woodford's 2005 paper on communication: "by making public its goals and methods, it increases its ability to conduct successful stabilization policy." It also holds the two halves of the record together without resolving them, naming forward guidance as an example of that transparency in its section 5, listing the forward-guidance puzzle among the framework's unresolved problems in its section 4.5, and citing none of Woodford's own papers on guidance.25 After 2008, when policy rates hit zero across the advanced economies, the Eggertsson and Woodford result was the warrant for quantitative easing and for the promises that accompanied it. Woodford's own verdict on that episode, given the day he won: forward guidance "did contribute to better stabilisation. It didn't magically solve the problem, but I think it was an important contribution."26
The framework's hardest test is now running at the Federal Reserve, where Chair Kevin Warsh has dropped forward guidance, declined to submit an interest-rate projection, and cut the Committee's statement roughly in half.27 The Delphic and Odyssean distinction is exactly what is at issue. Warsh has not gone silent, and in September he named a standard and four conditions under it, which is a reaction function in words with no number in it.28 Beth Hammack, speaking on the day of the prize, told business economists that "a reaction function is not forward guidance," and that treating them as interchangeable "can turn a well-meaning explanation of logic into the expectations of a promise."29 Markets have charged little for the change so far, with 10-year volatility inside its recent range and forecast dispersion below late 2025.30
Today's Fed has published nothing on the lower bound, where the theory makes its sharpest claim. Warsh's keynote never mentions the lower bound, the balance sheet, or asset purchases, and the balance sheet itself is one of five questions he has handed to an external task force that reports later.31 That omission answers to Eggertsson's later work rather than to the 2003 paper. With Saroj Bhattarai and Bulat Gafarov, he showed that quantitative easing works at the bound because holding long-duration assets makes future tightening expensive for the bank itself, generating "expectations of future monetary expansion in a time-consistent equilibrium."32 A balance sheet, on that reading, is how a central bank commits without having to promise. Returning to the bound is a tail risk, which the San Francisco Fed's market-implied estimate put at 5.6 percent over five years, and a plan for it costs nothing until it is needed.33
Asked in his Nobel interview what he admired in a central bank, Woodford named Sweden's Riksbank: its openness about policy, its willingness to explain outcomes that departed from its own forecasts, and its practice of talking "in advance about how things would most likely be done in the future."34 He said he thought everyone had a lot to learn from it, and that it was taken "more seriously perhaps in Sweden than even in my own country." He did not mention the Federal Reserve. Hall gave no interview at all, and Stanford's announcement of his prize does not quote him once.35