Keynes was the greatest British thinker of the 20th century. He had an extraordinary mind. He was a brilliant polemicist. And I am proud that he served in Her Majesty’s Treasury, whose view I shall seek to represent here tonight.
But 80 years on from the General Theory he remains elusive. Partly because he was a creature of his time - the chronic under demand of the inter war years. And partly because his ideas were continually evolving. Keynes was and is a paradox. A liberal who proposed protection. A capitalist who regarded most business people with contempt. A conscientious objector who worked round the clock in support of the war effort.
The General Theory is a masterpiece. It put macroeconomic analysis and policy firmly on the map. It provides huge insights into expectations, uncertainty and the operation of markets. His description of the stock market in chapter 12 should be compulsory reading for economists and investors alike.
It also provided much though not all of the basis for what came to be known as “Keynesianism”: a view that government could not just manage demand but seek to smooth the operation of the trade cycle through fiscal policy.
Whether Keynes himself would have supported such an approach, had he lived, we will never know: the General Theory was focused on addressing persistent depression. Chapter 22 (Notes on the Trade cycle) is almost an after-thought. It was Hicks, Meade and others who sought to operationalise “Keynesianism”.
Now is not the time to set out a defence of the much maligned Treasury view of the 1920s and 1930s. I would merely make two points.
First, the Treasury view evolved over time: as George Peden (1) has shown, it was much more nuanced than some of its critics have claimed. And secondly its focus on monetary policy as a way of regulating the economy, set out in Ralph Hawtrey’s seminal Economica article of 1925 (2), is still relevant today.
The Treasury policy of loose monetary policy and tight fiscal policy after the UK came off the gold standard in 1931 proved highly effective.
Similarly, in recent years, the speed of the authorities’ interventions on monetary and credit policy have been instrumental in the UK’s recovery.
And so the question I would like to address tonight is whether, beyond the initial loosening and tightening of fiscal policy by the then Chancellor in 2008-09, the Treasury should have made more use of Keynesian policies in recent years.
I will set out 9 reasons why the Treasury remains cautious if not sceptical about an activist fiscal policy. For completeness, I should make clear that many of the arguments apply equally to using monetary policy as a tool for fine tuning: the Treasury has always been as sceptical about crude monetarism as naïve Keynesianism.
First, the labour market is much more efficient than it was in the inter war period. Policy since the 1980s has focused on reforming industrial relations, improving work incentives and pursuing more activist welfare to work policies. Just as unemployment peaked at a lower level in the 1990s recession, so did it again in the 2009 recession, with unprecedented real wage adjustment facilitating the maintenance of employment. Keynes’ case for public works in the 1930s rested on his view that nominal (and hence real) wages could not adjust not least because of the strength of the trades union movement.
Secondly, over my working life, there has been a persistent tendency to mistake structural weakness for cyclical weakness. Keynes was writing at a time of chronically low demand but it’s not at all clear that recent experience fits this description. Apart from a brief hiatus in 2011 caused by the Eurozone crisis, unemployment has been falling persistently since early 2010: in the last three years, it has fallen by over a third, while the rate of employment has reached a record high. Throughout this period, until input prices began to plummet in 2014, core CPI inflation remained above its pre-crisis average, and did not fall below 2 per cent on a sustained basis until September 2014. Neither of these indicators are obvious signs of chronic lack of demand, and I doubt Keynes would have seen them as such, while the evidence is building that the growth of productive potential in the UK (and the US) has slowed significantly since the financial crisis. But throughout this period the “Keynesian” prescription has been the same: more stimulus and a higher deficit.
That naturally leads on to my third argument: the issue of asymmetry. For most of the post war period, Governments found it much easier to lower interest rates than to increase them, and to relax fiscal policy than to tighten it. No wonder there was a tendency for inflation always to be a little higher than desirable and for deficits to predominate at the expense of surpluses. Now, Gordon Brown dealt with the former through making the Bank of England operationally independent in 1997. In a democracy, it is difficult to see how fiscal policy could be contracted to an independent body. However, successive governments have sought to address this tendency through elaborate fiscal rules, and more recently through George Osborne’s creation of an independent Office for Budgetary Responsibility.
Fourthly, “Keynesian” demand management is likely to be much more effective in a relatively closed economy, like the United States, than an open economy like the UK. Here, demand expansion has historically fed through into imports and the current account: as Mark Carney recently pointed out, you cannot always rely on “the kindness of strangers” to help solve balance of payment problems. That led Keynes to argue for protection in the 1930s, just as Wynne Godley and the new Cambridge School argued for import controls in the 1970s. The Treasury has consistently set a very high bar when considering protection. Its commitment to Free Trade dates back to Gladstone. And you only have to look at the famous Kindleberger spider web diagram to see the damage protection did to the world economy in the 1930s.
Fifthly, the mythical “shovel ready” infrastructure project is precisely that: a myth. This is nothing new. The Treasury made the same point in the 1930s. But it is more of a problem today given the inexorable growth in planning law and wider regulation. Keynes’s suggestion in Chapter 10 of the General Theory that “the Treasury fill old bottles with bank-notes, bury them…in disused coalmines which are then filled up to the surface with town rubbish, and leave it to private enterprise…to dig up” would be the victim of many a health and safety regulation and environmental impact assessment today. In short, the lead times for getting public investment up and running are long and variable.
That leads some latter day Keynesians to advocate short term tax changes. Here, there tend to be administrative lags: for example, a change in the national insurance rate takes six months. That takes you inexorably to changes in VAT, which Alistair Darling reduced on a temporary basis in November 2008: that did bring forward expenditure albeit at some cost. An alternative is to increase current spending. But the problem there is that you can only do that by increasing entitlements, or employment or wages. Such changes are notoriously difficult to reverse.
Sixthly, there are the economic costs to businesses and individuals of continually changing tax rates and spending programmes. Businesses and consumers want a stable tax system. It enables them to plan with certainty. Tax policy is best set in a medium term framework, as for example the current Chancellor has been seeking to do with the corporate tax regime. The move to multi-year spending reviews from 1998 onwards also reflected the view of successive Chancellors that public service managers can spend money more efficiently if there is budget certainty over the medium term.
Seventhly, Ricardian equivalence is also relevant to fiscal policy’s effectiveness. A “permanent” stimulus will lead consumers to conclude that it will have to be financed, neutralising its impact. A theoretical case can be made that any Ricardian offset will be smaller if consumers know that a stimulus is temporary. Nevertheless, they are still likely to “look through” the change to some degree, reducing any inter-temporal effect. Whether for Ricardian reasons or because of wider leakages to imports, the Office for Budget Responsibility has estimated the fiscal multiplier at less than one. And interestingly, Nick Crafts(3) has estimated that fiscal interventions in the early 1930s would not have paid for themselves(4).
The role of the multiplier takes me to a eighth argument: the sheer magnitude of the fiscal interventions that would be necessary to stabilise the economy. This can best be illustrated either by looking at the extent to which the private sector savings ratio varies year by year; or by the extent to which output diverges from trend. The latter is much easier to estimate ex post than ex ante. But on the face of it, output has diverged from trend by up to 4 per cent of GDP since 1990. Using OBR estimates of the multiplier, stabilising the output gap would have required at times interventions of £100 to £250 billion compared to a neutral stance. And even if a limit was placed on discretionary counter cyclical interventions of, say, 1 per cent of GDP in any one year, there would still be regular changes in policy of up to £18 billion a year. Whether or not that would unsettle the market, it would certainly trigger the damaging effect on economic efficiency I mentioned above.
Finally, I would argue that there are positive benefits (as well as costs) to the trade cycle provided it can be kept within reasonable bounds. As Nigel Lawson has said, “the superiority of market capitalism lies in particular in two areas: the freedom and encouragement it gives to innovation and risk taking…,and the discipline that drives up efficiency and drives down costs. The former is stimulated most during the cyclical upswing, and the latter is compelled most during the downswing. It is at least arguable that if economies moved in a straight line rather than a cyclical pattern, there might, in the long run be less of both these benefits.(5)” In short, Schumpeter may still have as much relevance today as Keynes.
The Treasury may be sceptical about activist demand management. But that does not mean it abdicates responsibility for economic performance. As the nation’s economics ministry, it attaches a high weight to microeconomic policies that promote growth, productivity and employment.
Since the 1970s, successive Chancellors have sought to create a macroeconomic framework which seeks to create price stability. For the most part the Treasury has relied on monetary policy to achieve that objective: since 1997, an operationally independent Bank of England has been tasked with hitting a symmetric inflation target. Fiscal policy has generally played a subsidiary role, with Chancellors setting it to achieve a medium term objective – a surplus under Nigel Lawson and George Osborne; a current surplus under Gordon Brown – underpinned with a target for the national debt.
That does not mean that there is no role for fiscal policy. In the recent downturn, the automatic stabilisers played an important role in supporting demand. As George Osborne has said “by not chasing the debt target we…allowed the automatic stabilisers to operate and that is a sensible economic decision… That supports the economy in that sense, during a cyclical downturn.(6)”
And as a pragmatic institution, the Treasury would never rule out recommending a fiscal response if the conditions were right.
But it is no surprise to me that the response to the recent crisis has focused on monetary policy and the credit channel rather than on fiscal policy. In 2008 we saw the advent of the special liquidity scheme, the credit guarantee scheme and “quantitative easing”. Latterly, we have seen the funding for lending scheme, supplemented by other interventions such as “help to buy” – all of which have been designed to reduce the gap between official interest rates and the rates companies and households pay.
If you have a banking crisis followed by a credit crunch, you need to treat the disease rather than the symptom. Similarly, it’s in the nature of a banking crisis that government deficits are likely to rise, often sharply. That is not a time to take risks with the deficit - there are always inflection points when just a little extra borrowing can do untold damage to how you are perceived in the market. Yields start to rise. Debt servicing costs begin to spiral. And that risk increases if you go into a down turn with an already high debt level.
Some neo-Keynesians may write off the modern Treasury view as expounded by “practical men who believe themselves to be quite exempt from any intellectual influence, [but] are usually the slaves of some defunct economist. [Or] Madmen in authority, who hear voices in the air, [and] are distilling their frenzy from some academic scribbler of a few years back.(7)” But I’d like to think that Maynard Keynes – who understood markets as well as anybody – would have approved of what the Treasury has done since 2008.
- G C Peden (2004), Keynes and his critics, Treasury Responses to the Keynesian Revolution 1925-46
- “The true remedy for unemployment is to be found in a direct regulation of credit on sound lines” R. G. Hawtrey (1925), “Public Expenditure and the Demand for Labour,” Economica 13
- Nick Crafts and Terence Mills (2015): Self-defeating austerity? Evidence from 1930s Britain, European Journal of Economic History, 19. Interestingly, they conclude that in the later 1930s there was a fiscal free lunch
- Perhaps, this explains P J Grigg’s conclusion that he could never see in Keynes’ various proposals “anything but different manifestations of a thesis that nations, unlike private individuals, can live so as both to have their cake and eat it”. (GC Peden, The Treasury View in the interwar period”, in Unemployment and the Economists, Corry, 1996). Grigg was Principal Private Secretary to the Chancellor from 1921 to 1930.
- Nigel Lawson, Memoirs of a Tory Radical (Revised edition, 2010)
- Oral evidence to Treasury Committee, 13 December 2012
- J M Keynes (1936), The General Theory of Employment, Interest and Money, Chapter 24, Concluding Notes