Rebuilding after the Second World War: what lessons for today?

The UK’s ratio of public debt to GDP was very big after the Second World War, but economic conditions were highly favourable for its relatively painless reduction. Interest rates and unemployment were low, and economic growth was rapid. This meant that it was not necessary to run budget surpluses to bring the debt ratio down. Expenditure on the early post-war welfare state was quite modest by later standards – and the taxes to pay for it could be levied quite readily in the aftermath of the war.

Reducing the ratio of public debt to GDP

At the end of the war, net public debt to GDP peaked at 252% in 1946/47; 25 years later in 1971/72, it was only 62.1%. This large fall was achieved despite government budget deficits that averaged 1.2% of GDP during the 1950s and 1960s.

A little bit of budgetary arithmetic explains how this came about. A well-known formula tells us how big is the primary budget surplus as a percentage of GDP, b*, which is required to prevent the debt ratio, d, from rising:

b* = d(r – g)

Where r is the real interest rate (the interest rate on government debt minus the rate of inflation) and g is the rate of growth of real GDP. The primary budget surplus is government outlays excluding debt service minus government receipts. This formula shows that if the real interest rate is less than the growth rate, it is possible to reduce the debt ratio while running a primary budget deficit.

In the 1950s and 1960s, the primary budget surplus averaged about 2.3% of GDP, the real interest rate averaged 0.2%, and the growth rate averaged 3.0%. This implies that about 60% of the reduction in the debt ratio came from the excess of the growth rate over the real interest rate and only about 40% from the primary budget surplus (Crafts, 2016).

Why was ‘r minus g’ so favourable?

The low real interest rate reflected an era of ‘financial repression’ underpinned by capital controls that stopped an outflow of savings to other countries and onerous bank regulations that compelled banks to hold large amounts of government debt (Allen, 2014). This allowed the government to borrow from ‘captive lenders’ on very favourable terms, which would not have been available in a free market environment.

The strong growth rate reflected an era in which European countries generally benefited from rapid growth as they reduced the initially huge productivity gap with the United States at a time of of high investment, rapid technological progress, restructuring and economic integration. In fact, the UK’s performance was comparatively disappointing but nevertheless good enough for pleasant fiscal arithmetic.

Expansion of the welfare state

The early post-war years saw the implementation of much of the Beveridge Report on social security during 1946 to 1948 and the establishment of the National Health Service in 1948. Government spending on social security as a percentage of GDP did not increase initially – it was 4.9% of GDP both in 1938 and in 1951, but it was 8.3% in 1974 at the end of the ‘golden age’ of economic growth.

In the late 1940s, much higher expenditure on pensions than before the war was largely offset by lower outlays on unemployment benefits. Health expenditures rose from 1.6% of GDP in 1938 to 3.4% in 1951, and then to 4.6% by 1974 (Middleton, 1996).

Taxes (including national insurance contributions) as a percentage of GDP rose markedly from 21.6% in 1937 to 33.5% in 1951. The main burden of increased taxation fell on income taxes, which rose from 6.2% of GDP in 1937 to 13.4% in 1951.

Taxes relative to incomes rose for all income levels but especially for the well-off. For example, in the late 1940s a household earning £3-5,000 per year would have had a total tax burden of about 80%, roughly double what it would have paid in taxes pre-war. It is generally agreed that both world wars had a ‘displacement effect’, which significantly ratcheted up tolerance for higher taxation.

Were the objectives of the Beveridge Report achieved?

The Beveridge Report had the objectives of bringing everybody up to an acceptable minimum level of income by providing adequate benefits and of abolishing means testing. Neither was achieved.

Although in the early 1950s it was commonly believed that poverty had been abolished, this was not the case. Modern analysis of the data from Rowntree’s survey of York in 1950 shows that 11.8% of working class households (and 7.1% of all households) were below his poverty line (Hatton and Bailey, 2000).

Using a modern poverty standard of 60% of median income, Gazeley et al (2017) find that 13.2% of all households were in poverty in 1953/54 according to the family expenditure survey, compared to perhaps 16% in the late 1930s. Many people continued to need National Assistance, which was intended to be a means-tested safety net: there were one million claimants in 1948 and two million by the mid-1960s.

There were several reasons why the welfare state reforms ‘failed’, including fundamental flaws in its design as a so-called ‘social insurance’ scheme based on flat-rate national insurance contributions (Whiteside, 2014). It must be accepted, however, that a key reason was that benefit levels were set too low to provide the ‘adequacy’ to which Beveridge aspired.

Was the full Beveridge compatible with reducing the debt ratio?

The answer is clearly ‘Yes’!

If the aim was to achieve Beveridge’s objectives using his basic design and accepting Rowntree’s poverty line, expenditure on social security would have been about 35% (1.7% of GDP) higher based on a study by Dilnot et al (1984). Atkinson (1969) came to a similar conclusion, namely that the cost of a ‘back-to-Beveridge’ set of reforms in the late 1960s was about 1.7% of GDP.

Given the fiscal configuration of the 1950s, the debt ratio would have been falling provided that the primary budget deficit was no greater than 5.6% of GDP. In fact, there was a primary budget surplus of 2.6% of GDP on average in those years.

So, with hindsight, there was plenty of scope to choose a somewhat slower path for reduction of the debt ratio while spending enough on welfare to meet Beveridge’s original ambitions by borrowing a bit more, as long as there were no adverse fiscal shocks, such as a return to the high unemployment of the 1930s – an outcome that had been much feared at the end of the war.

Is r likely to be less than g in future?

This is a key question whose importance is often not well recognised in discussions of fiscal sustainability and of the scope to reach a new social settlement post-coronavirus. It was absolutely central to expanding the welfare state at a time when the debt ratio was uncomfortably high and needed to be brought down after the Second World War.

The Office for Budget Responsibility (OBR, 2019, chapter 7) does, however, consider the issue very thoroughly. Its pre-crisis forecast was that ‘r minus g’ might average about -1.1% over the next five years. The OBR’s current assumption is that net debt will be close to 100% of GDP at the end of fiscal year 2020/21, so a modest primary deficit would be feasible while stabilising the debt ratio at its new inflated value.

The future of ‘r minus g’ is unclear, especially for g. Given future demographics, the trend growth of real GDP will almost entirely come from labour productivity growth rather than increases in employment. Unfortunately, recent UK productivity growth has been unprecedently bad (Crafts and Mills, 2020). In the absence of an improvement in productivity performance that raises g, it is not difficult to imagine that ‘r minus g’ could flip to positive, especially if markets start to see UK government debt as increasingly risky (Wyplosz, 2019).

Where can I find out more?

Relevant statistical data on public finances can be found in the Office for Budget Responsibility’s PSF Aggregates Databank.

Statistical information on historical social security spending can be found on the Institute for Fiscal Studies website.

Office for Budget Responsibility, Fiscal Risks Report July 2019 contains a detailed discussion of the impact of interest rates and growth rates on reduction of the ratio of public debt to GDP.

Origins and measurement of financial repression: the British case in the mid-20th century: Garrick Hileman explains the background to control of interest rates after the Second World War.

The Beveridge Report and its implementation: a revolutionary project? Noel Whiteside provides an outline and critique of the Beveridge Report.

Economic lessons of 1945: BBC Radio 4 discussion on Evan Davis’s The Bottom Line programme with (economic) historians Margaret MacMillan, Catherine Schenk, David Edgerton and Jim Tomlinson.

Who are UK experts on this question?

• Nicholas Crafts, University of Sussex and CAGE, University of Warwick.
• William Allen, NIESR has worked on financial repression in the UK after the Second World War.
• Ian Gazeley, London School of Economics works on poverty and inequality in the UK during the 20th century.
• Roger Middleton, University of Bristol has worked on the history of government spending.
• Andrew Scott, London Business School works on the history of the national debt.
• Pat Thane, King’s College, London has worked on the history of the welfare state in the UK.

Author: Nicholas Crafts (University of Sussex and CAGE, University of Warwick)