Keynesian in a Foxhole

On​ 14 November 1968, New York University hosted a debate between the economists Walter Heller and Milton Friedman at the Graduate School of Business, a block from Wall Street. The event was advertised as a confrontation between two men engaged in a war for the soul of American economic policy. In retrospect, it might not sound like a clash of the titans – Milton Friedman versus who? – but in 1968 Walter Heller was a well-known figure. The hall, though large, wasn’t big enough to seat the hundreds of balding, besuited businessmen who showed up; closed-circuit televisions had to be installed in adjacent classrooms. Heller was the era’s most influential proponent of Keynesianism, then the dominant approach to economics in the United States. Unlike Friedman, he was much better known for his government work than his scholarship: he led the Council of Economic Advisers from 1961 to 1964, under Kennedy then Johnson, and had been economic adviser to Hubert Humphrey’s presidential campaign until, just a week before the debate in New York, it ended with the election of Richard Nixon. Friedman had been closely involved with Barry Goldwater’s reactionary, anti-desegregationist campaign in 1964, and was now an adviser to Nixon; the New York Times called him a ‘radical conservative’. He was also the foremost advocate of monetarism, an avowedly anti-Keynesian approach to macroeconomics.

The two main tools used by modern capitalist states to run the economy are fiscal policy (taxing and spending) and monetary policy (regulating the circulation of credit and debt in the financial system). The debate between Friedman and Heller concerned the power and limits of these tools, a question that continues to trouble us today. To frame the discussion, each of them was assigned a question. For Heller: ‘Is monetary policy being oversold?’ For Friedman: ‘Has fiscal policy been oversold?’ The difference in verb tenses indicates what the audience already knew: at that time, for the most part, fiscal policy called the tune and monetary policy danced to it. Naturally, both men answered ‘yes’ to their question. As a ‘Keynesian’, a term by then associated with the use of policy to ‘fine-tune’ the economy, Heller was an advocate of co-ordination between fiscal and monetary policy, and of giving the state ‘discretionary’ power. Friedman, a relentless proponent of small government and laissez-faire, had the year before used his presidential address to the American Economic Association to argue that economies have a ‘natural rate of unemployment’ at any given moment. It is baked into the structure: you can try to lower it, but you will fail. This means that expansionary policy (policy intended to stimulate growth and employment) is inflationary and, in anything but the short run, useless or worse than useless – a conclusion which, depending on your point of view, was either proof of, or predetermined by, the monetarist position.

The term ‘monetarist’ was coined by Karl Brunner, an ally of Friedman’s, in an article published in the St Louis Federal Reserve’s Review in July 1968; by November it was already in widespread use. Today the term is generally used to describe a ‘conservative’ or even ‘right-wing’ approach to macroeconomics, but beyond that the definition gets quite fuzzy. What Friedman and his fellow monetarists mean by it is more precise: not just something to do with money, but specifically to do with the nominal money stock, that is, the actual quantity of money circulating in the economy, including money created by the central bank, credit created by the banking system, and deposits and currency held by the public. To say, as Friedman did in 1970, that ‘inflation is always and everywhere a monetary phenomenon’ is to say that inflation is always and everywhere a result of growth in the money stock unmatched by growth in output. ‘Too much money chasing too few goods’, as the saying goes. The policy implication is that if we control the growth of the money stock, we can – ‘always and everywhere’ – control inflation.

Friedman claimed that the historical evidence bore him out. In a monumental book from 1963, A Monetary History of the United States, 1867-1960, he and Anna Schwartz argued that across an extraordinary period of economic transformation, changes in the money stock had been ‘closely associated with changes in economic activity, money income and prices’, and that these relations were ‘highly stable’. The success of the ‘Keynesian revolution in academic economic thought’ had, in their view, led economists to conclude that ‘money does not matter, that the stock of money adapted itself passively to economic changes and played a negligible independent role,’ when in fact the causal forces ran in the other direction: economic activity responds to monetary change, not the other way around. The Great Depression in the US would have been nowhere near as severe, they claimed, and capitalism would have righted itself, if only the Federal Reserve had prevented a catastrophic contraction of the money stock. In their telling, history showed that the private market economy as a whole was basically stable, and the often disastrous volatility of capitalist economies wasn’t a reason to deploy government stabilisation policy, but the result of its deployment.

Monetarists drew two main lessons from this: first, that the money stock was the one and only thermostat that policymakers required to regulate an effectively self-correcting economy; second, that rules should be introduced, via what Friedman called ‘constitutional provision’, to insulate monetary governance from the temptation to meddle with an economy that was better left to its own devices. If everyone expected that every year the money stock would increase by the same amount as the anticipated rate of economic growth, and the government had to follow through on that expectation, the world would be an anti-inflationary utopia.

Keynesians like Heller thought that history taught exactly the opposite lesson. ‘The issue,’ he told the audience in New York, ‘is not whether money matters – we all grant that – but whether only money matters, as some Friedmanites, or perhaps I should say Friedmanics, would put it.’ As he saw it, ‘while the monetarists tell the policymaker, in effect, “Don’t do something, just stand there,”’ the whole postwar experience was evidence that we should instead rely on ‘the economic wisdom, the strength of character and the institutional capacity to operate a successfully discretionary policy’. The last thing the economy needed was ‘rigid and static rules’: ‘Let’s not lock the steering wheel into place, knowing full well the twists and turns in the road ahead. That’s an invitation to chaos.’

Alan Blinder’s A Fiscal and Monetary History of the United States, 1961-2021 is the latest contribution to what is in many ways the same argument that Heller and Friedman were having more than fifty years ago. What should we expect of economic policy? What can it do and what are its limits? Should one mode of policymaking ‘dominate’ the other? (Friedman the monetarist abhorred ‘extreme fiscalists’.) Blinder is well placed to attempt some answers. He is an insider: an eminent academic economist, based at Princeton, who has been deeply involved in both fiscal and monetary policymaking, first as a member of Bill Clinton’s Council of Economic Advisers and later as vice chairman of the Federal Reserve. But despite the allusion to Friedman and Schwartz in the book’s title, he is decidedly not a monetarist, but a Keynesian, a regular contributor to the Wall Street Post and a self-described ‘centre-left Democrat’. He is also an engaging writer about ideas and processes that some economists purposefully obfuscate or consider the public too stupid to understand. The relationship between democracy and economic governance isn’t always happy on Blinder’s account – and often he suggests it is democracy that has to give way – but he is clearly dedicated to a more democratic distribution of economic knowledge.

Blinder’s economic views are generally of the sort labelled ‘new Keynesian’ – one of the dominant modes of contemporary mainstream economics. It is ‘new’ in so far as it is the product of several compromises Keynesians have made over the last half-century both with neoclassical economics and with monetarism and its descendants, regarding two things in particular: the kind of ‘rationality’ that motivates economic activity, and the limits on policy effectiveness. Like all varieties of Keynesianism, it begins from the premise that the modern capitalist economy is, as Keynes himself put it, a ‘delicate machine’, prone to misfire and breakdown. In the words of Franco Modigliani, a prominent Keynesian of Friedman’s generation, this means that ‘a private economy using an intangible money needs to be stabilised, can be stabilised and therefore should be stabilised by appropriate monetary and fiscal policies.’

Modigliani’s imperatives are grounded in a fundamental conclusion of Keynesian analysis: that laissez-faire tends towards disorder and instability. Capitalist economies, left to themselves, will not ‘naturally’ employ all the resources available. Markets will not automatically ‘clear’ and some resources, including people looking for work, will remain idle despite their productive potential. Sometimes, as during the Depression, when Keynes was writing, the economy can leave an enormous number of people behind. These people are unemployed involuntarily – meaning they would take a job at the going rate if one were available – and that is not their fault. The state must help, and if it does, it will benefit everyone because capitalism is driven by demand. The more people working, spending and investing, the better for all.

In the wake of crises like the financial meltdown of 2008 and the Covid-19 pandemic, these propositions might seem uncontroversial. But monetarism broke the policymaking surface just as the Keynesian promises of the 1960s were turning into the perils of the 1970s. The inflation-unemployment spiral or ‘stagflation’ that seized the wealthy, industrialised part of the world in the 1970s seemed simultaneously to give the lie to Heller’s faith in discretionary ‘economic wisdom’ and to suggest that expansionary stabilisation policy was indeed making things worse. Blinder argues persuasively that Keynesianism was ‘unjustly’ tagged as inflationary at the time; he and others, in the 1970s and since, have explained stagflation as a product of supply shocks, especially in food and energy. But monetarist common sense – which Modigliani summed up as ‘the view that there is no serious need to stabilise the economy’ – started to enjoy mainstream support. In 1977, Congress directed the Fed to begin tracking monetary aggregates (different measures of the money stock) for the purposes of monetary policy, and then, in August 1979, President Jimmy Carter appointed Paul Volcker as chairman of the Fed.

When Volcker was appointed, the federal funds rate (the Fed’s primary policy tool, the rate at which banks lend to each other) was 11 per cent; by June 1981 it was 22 per cent. This was the ‘Volcker shock’, which crushed inflation by effectively shutting down borrowing, spending and investment. The event is held up by some as monetarism’s moment of triumph: Volcker, they say, had tamed the discretionary Keynesian monster. As Blinder sees it, this is baloney. When Volcker took over the Fed, the monetarists’ stable relationship between money and economic activity was nowhere to be found. He did not generate the anti-inflationary recession of the early 1980s through a vigilant targeting of money growth, which proved to be both a bad idea and operationally impossible. Indeed, as Blinder puts it, by the autumn of 1982, ‘monetarist doctrine stood in the way.’ It ‘had to go’ and it ‘had not worked anyway’. Instead, Blinder says, Volcker used monetarist doctrine as a ‘convenient political heat shield’ for ‘excruciatingly high interest rates’. Volcker himself later said that the simplicity of Friedman’s ‘always and everywhere’ dictum ‘helped provide a basis for presenting the new approach to the American public’.

Blinder can’t contain his schadenfreude at the disintegration of the monetarist experiment, and turns his gaze on the incoherence that followed it in the shape of Reagan-era ‘supply-side economics’. Reaganomics excoriated Keynesianism while generating some of the biggest budget deficits in American history, owed largely to defence spending. Looking back, the moment might seem to have offered an opportunity to reassert what Blinder would consider economic common sense: a ‘two-sided’ Keynesianism in which stabilisation policy gives the economy a kick when it is slowing, and reins it in when it starts speeding up. Reagan’s policy was in fact a one-sided Cold War ‘military Keynesianism’ pretending to be hard-nosed conservative economics.

But in both academic and policy circles, Keynesianism hadn’t recovered from the beating it took in the 1970s. An aggressively anti-Keynesian post-monetarism began to dominate macroeconomics, dedicated to monetarism’s free-market fundamentalism but virtually uninterested in the money stock, in which it placed no faith and barely even mentioned. The first stirrings of this self-described ‘rational expectations revolution’ began with the work of Friedman’s student Robert Lucas. Keynes had put expectations at the centre of macroeconomics. If you wanted to understand how an actually existing market economy worked, you had to begin with what people expected the future would look like, because those expectations determine their behaviour. Pretty much every economist agreed with this. But the monetarists bristled at how ‘irrational’ those expectations seemed in Keynesian thinking, which in their view depicted market actors as myopic, impulsive, gullible and sometimes just dumb. This undermined the entirety of neoclassical economics, based as it is on rational, self-interested decision-making by optimising agents operating in complete markets (conditions in which all goods are available and subject to competition).

Lucas’s ‘new classical economics’ was an attempt to end this sacrilege. It was built on models with neoclassical ‘microfoundations’, which assumed that the expectations of market participants are fully ‘rational’: that the participants have all available information, including the information policymakers have, and use it just as ‘efficiently’; that they know and respond to the ways policymakers operate; and that they know what their lives will be like, so that every decision moves optimally towards that end. This work was inspired by Friedman’s effort to demonstrate the ultimate futility of stabilisation policy. If actors have rational expectations, they know what is probably going to happen in the economy, so can anticipate it. This means ‘systematic’ policy is totally ineffective. The only way to ‘stimulate’, in that case, is to ‘surprise’. This might work in the very short term, but soon it will just end up generating inflation that will be expected, and then the only way to ‘surprise’ anyone will be an even bigger surprise, causing even more inflation. Everything the government does to ‘help’ will end up being irrelevant, leaving the economy back where it started but with ever higher prices.

Lucas, whose new classical economics seeded further uncompromising variants such as ‘real business cycle’ theory and ‘dynamic stochastic general equilibrium’ macro, insisted that the only meaningful economics is technical mathematical modelling: ‘Everything else is just pictures and talk.’ Keynes’s General Theory of Employment, Interest and Money (1936), the book that underwrote the Keynesian project, was nothing more than ‘disconnected qualitative talk’. Further, all macroeconomic models had to have the neoclassical ‘microfoundations’ that drive the rational expectations revolution. Nothing else was ‘real social science’. In the labour market, all workers (‘labour suppliers’) are rationally optimising the ‘intertemporal substitution’ of work and leisure across their lifetimes: as Lucas put it, ‘To explain why people allocate time to a particular activity – like unemployment – we need to know why they prefer it to all other available activities: to say that I am allergic to strawberries does not “explain” why I drink coffee.’ (No, I don’t get the metaphor either.)

Finally, all macroeconomic models must begin from the premise that the economy as a whole is always in intertemporal equilibrium: all markets ‘clear’ (all supply finds demand), meaning ‘idle’ resources are not proof of states of disequilibrium, but rather of the unfolding process of rational self-interested decision-making. This follows directly from the ‘microfoundations’ assumption: if the people that make up an economy are rational agents optimising intertemporally across their lifetimes – or even beyond, into overlapping generations – then at no point could that economy possibly be in disequilibrium. Everything is as it is because of decisions made to maximise agents’ utility over their life-cycles. The operation of the economy isn’t actually a ‘macro’ phenomenon but the product of ‘aggregative’, rational, individual choices; disequilibrium is thinkable only on the Keynesian assumption of what Lucas once called ‘unintelligent behaviour’.

The upshot is the polar opposite of Keynesianism: markets produce the optimal outcome; the state is the problem, not the solution. On this account (which Blinder thinks amounts to a sort of ‘econometric nihilism’), Modigliani was perfectly wrong on all counts: capitalist economies do not need to be stabilised, and they cannot and should not be stabilised by monetary and fiscal policies. What’s more, there is no such thing as involuntary unemployment, the foundation of Keynesian politics and policy. Indeed, there is no such thing as unemployment at all: it is an ‘activity’ to which people ‘allocate time’, essentially an ‘ideological’ way of saying ‘leisure’.

The monetarist project and its Lucasian variants amount to nothing less than the ‘scientific’ delegitimisation of the welfare state. They are an attack on the very idea of the collective, even its frayed remaining strands, and – this is not an exaggeration – provide the operational framework for Thatcher’s claim that ‘there is no such thing as society.’ The unwillingness of its advocates to acknowledge this is unsurprising: this is the economics of a powerful elite, insulated from hardship, poverty and ill fortune. Their analysis is conducted without reference to, let alone consideration for, the circumstances and decision-making of the real people who feature as the stick-figure unemployed in their models.

And yet the proponents of new classical economics and more recent rational-expectations approaches parrot Friedman in the adolescent disavowal of their political motivations, which are, as the New York Times in November 1968 could plainly see, ‘radically conservative’. Friedman, whose anti-desegregation, pro-Pinochet positions were notorious, opened his contribution to the debate with Heller by assuring the audience that the discussion would have ‘no political element whatsoever’. He didn’t mean it as a joke. Lucas, too, denies any ‘simple connection’ between his economics and his political commitment to ‘limited government, budget balance and tight money’, despite the fact that they are always in sync.

Blinder writes that the evidence for monetarism was never ‘very convincing – unless you were already convinced’. (He has been making these criticisms for decades. In 1988, he called new classical economics ‘a triumph of a priori theorising over empiricism, of intellectual aesthetics over observation and, in some measure, of conservative ideology over liberalism’ – ‘liberalism’ in the American, social-programmes-are-good sense.) He tries to resist debating Friedman, Lucas and the like, though sometimes he can’t help himself. What bothers him most, however, isn’t the narrow-mindedness or arrogance of post-monetarist free-market fundamentalism, or the spiralling inequality and environmental destruction it has legitimised and even celebrated. Blinder’s main complaint is that it hasn’t helped decision-makers, and indeed would have set policymaking back decades were it not for the fact that most people have ‘had the good sense to (mostly) ignore it’.

Blinder has been​ personally involved in American economic policy since the 1990s (many of the footnotes to the final chapters begin ‘Full disclosure’). The last thirty years have been a period of significant fluctuation in policy dominance. The Clinton administration’s obsession with deficit reduction, mainly via spending cuts, put fiscal policy in the driver’s seat. Alan Greenspan, the Fed chair between 1987 and 2006, was an Ayn Rand libertarian with strong Republican sympathies, yet during the Clinton years he declined to make significant policy moves because in his assessment the boom times and low unemployment weren’t inflationary but a product of increased labour productivity. Blinder gives him a lot of credit for his ‘forbearance’. The economic expansion that followed the dotcom ‘recessionette’ in the early 2000s was also an era of fiscal priority, as George W. Bush, when he was paying attention, focused on tax cuts for the rich and corporations, and on military spending for the destruction of Iraq. In this he was abetted by Greenspan, who publicly defended the tax cuts (a violation of central bankers’ political neutrality that greatly troubles Blinder) and drastically lowered interest rates. Indeed, he lowered them by so much that when the financial crisis exploded in 2007, he had little room left to lower them further.

Blinder has written extensively on the origins and fallout of the crisis. In A Monetary and Fiscal History he summarises his account in After the Music Stopped (2013), arguing that the crisis was the result of indefensible financial ‘shenanigans’ and a somewhat more defensible failure to regulate the financial sector. Blinder remains disgusted by the scheming and wilful blindness of the time – the details are ‘grotesque’, ‘ludicrous’, ‘hideous’ – but the regulatory response was, in his view, something to be celebrated, a series of creative, seat-of-the-pants experiments on the part of the Fed, which, recognising that it could respond to the situation much faster than Congress, stepped carefully into ‘quasi-fiscal’ territory and rescued the American economy.

The chairman during the crisis, Blinder’s Princeton colleague Ben Bernanke, is one of the book’s heroes. Another is Jerome ‘Jay’ Powell, who was appointed Fed chair by Trump – the only thing Trump did that Blinder thinks was a good idea. Powell, he writes, ‘had greatness thrust upon him’, leading the central bank through the stormy waters of the pandemic with creativity, flexibility and ‘strength of character’ in the face of Trump’s constant attacks. Blinder is big on ‘strength of character’: in his estimation Volcker, Greenspan and Bernanke had a lot of it too. By contrast, there aren’t a lot of fiscal policy heroes in A Monetary and Fiscal History. This is no doubt a reflection of Blinder’s own interests: monetary policy is his field and he has been a central banker himself. But the more fundamental reason that Blinder’s great ones are virtually all central bankers, not lawmakers, is his deeply held belief that ‘politics’ has no place in economics. Most modern central banks enjoy a legislated independence from elected officials, and as Blinder wrote as long ago as 1993, politicians tend to ‘reach for short-term gains at the expense of the future’, so we are wise to put monetary policy ‘in the hands of unelected technocrats with long terms of office and insulation from the hurly-burly of politics’.

Fiscal policymakers, though, cannot separate themselves from the hurly-burly. Political logic is expedient, compromising and often, on Blinder’s account, just plain irrational: ‘Logic is never required in a congressional hearing.’ Blinder, unlike the monetarists, isn’t so arrogant as to claim we need a ‘constitutional economics’, but he does share with them the belief that politics, even democracy itself, has an ‘inflationary bias’ that must be subdued. Both Blinder and his opponents on the right argue that central bank ‘transparency’, clear communication about its commitments and plans, will have to suffice as a substitute for democracy, which tends only to make things worse. He dismisses new classical economists’ claims about policy ineffectiveness, but is convinced that wherever economics is motivated by ‘politics’, it will do more harm than good.

This shared distrust hints at deeper commonalities between approaches. On Blinder’s account, it is always Keynesianism that saves the day. Not only do policymakers always ‘turn Keynesian very quickly’ in an emergency – as Lucas himself put it in 2008, ‘Everyone is a Keynesian in a foxhole’ – but good policy has often been Keynesian even while pretending it wasn’t: Greenspan, a vociferous anti-Keynesian, was a ‘consummate fine-tuner’. Blinder has long argued that good policymaking is both an art and a science, a mix of intuition and technical analysis. This is the mix that he thinks Keynesianism provides, and Keynesians, being open-minded, flexible thinkers, have absorbed what is ‘useful’ from anti-Keynesian approaches.

But this understates the extent to which recent new Keynesian economics has borrowed from monetarist and Lucasian thinking. As the Stanford economist John Taylor – who may not be Blinder’s kind of new Keynesian, but is a new Keynesian nonetheless – put it, ‘the rational expectations hypothesis is by far the most common expectations assumption used in macroeconomic research today.’ Taylor even claims that the origins of new Keynesianism can ‘be traced to the introduction of rational expectations into macroeconomics in the early 1970s’. Blinder is certainly right that, for example, modern macro models all contain ‘Keynesian’ features, such as the recognition that competition isn’t perfect and that prices don’t adjust instantaneously to market signals (they are ‘sticky’), but these features are superimposed on structures drawn directly from the new classical approach, like ‘rational microfoundations’, general equilibrium and market-clearing. Blinder often leans on Friedman’s notion of a ‘natural rate of unemployment’ below which we should not go, as if it were an incontestable fact of reality. Which, in mainstream economics, it is.

The ‘natural rate’ is one of those ideas that makes visible the politics of orthodox economics. What makes that particular level of unemployment ‘natural’ and, more important, why should we accept it as a given? Why is there no ‘natural rate’ of profit, poverty or inequality beyond which we shouldn’t go? To say that economics is political isn’t to say it is conniving or ill-willed, but merely that, like any form of knowledge aimed at understanding a complicated world, it is intensely social. It is produced by particular people in particular places at particular times. Economists’ posture of ‘objectivity’ is like an accent; since they mostly talk among themselves, they can’t hear it.

At the same time, it is of course possible for ideology to parade as science. Monetarism is a case in point. Friedman actively opposed the Civil Rights Act, saying racism should help black workers get jobs, because it made them cheaper to hire. Lucas claimed that his models help us understand the ‘situations people find themselves in, the options they face and the pros and cons as they themselves see them’, but in those models everyone is an identical, featureless rational optimiser; in fact, because everyone is identical, they are clumped into one ‘person’, the ‘representative agent’. It is hard to see how that sort of thinking will help anyone who needs help.

Blinder, though, wants to care, and he thinks Keynesianism is an even-handed, practical way of caring. He isn’t delusional or mendacious enough to put his ideas forward as immutable truth. He recognises that in the ‘real world’ politics and economics are inseparable, but he is a Keynesian in his commitment to building a wall between them, however artificial. He is less of a Keynesian than Keynes, however, in his belief that the world of ‘feasible’ economies contains only capitalism, and its economics only variations on mainstream analysis, albeit of the ‘centre-left’.

If we confine ourselves to the narrow corridor of economic possibility in which Blinder operates – Keynesianism v. anti-Keynesianism – then, sure, Keynesianism is better. But out in the real world, where the chief merit of Blinder’s Keynesianism is that the one alternative it can imagine, the only analytical opposition it even recognises, is a total disaster, this is hardly a reason to celebrate. Emergency measures are better than nothing, and it would be something of an improvement if we were Keynesians not just when we are trapped in foxholes. But as the natural and social fabric of collective life on earth frays and unravels, our way of thinking about economics needs a more fundamental re-examination.

Ironically, it is only politics that will turn economics around. Economists can and do help, and they need not be self-consciously ‘radical’ – the work of Jayati Ghosh and Carolina Alves comes to mind, as well as that of Martín Guzmán, until recently economic minister of Argentina. These are people who abandoned or never bothered in the first place with the pretence that economics is apolitical. They all see that what we have is clearly not working for many. The only response is to go further and wider, to abandon ‘the market’ as the fixed point of economic reference along with the walls of convention that surround it. I expect Blinder will think these the musings of a naive utopian. But the world can no longer live with an economics that confines itself to his numbed sense of the possible.