Osborne

G7 growth rates and austerity

Rob Calvert Jump and Jo Michell

In August 2022, revisions to official measures of UK output generated headlines because the new figures implied that the economic contraction during the pandemic was greater than previously thought. 

At the same time, however, substantial revisions were made to historical data, and these received far less attention. One outcome of these revisions is that the UK’s performance relative to other rich economies during the austerity period of 2010–2016 has been downgraded: growth in real GDP per capita over this period is now meaningfully lower. This means that some recent analyses relying on the older figures are misleading.

For example, in a recent FT article, Chris Giles includes data showing that the UK had the highest growth of real GDP per head in the G7 between 2010 and 2016. Inevitably, the article was circulated by defenders of austerity including Rupert Harrison and Tim Pitt, alongside a claim that the data “shows why the idea austerity has caused our growth problems post-GFC doesn’t stack up. During peak austerity (2010-6) UK had strongest GDP per capita growth in G7”.

The data used by Chris Giles are from the International Monetary Fund’s (IMF) October 2022 World Economic Outlook (WEO), and show average annual growth in real GDP per head of 1.4% in the UK between 2010 and 2016, compared with 1.3% in both Germany and the USA. But the October 2022 WEO uses data from the 2021 Blue Book, which were compiled before the most recent set of revisions were introduced.

The 2021 data imply that total per capita growth between 2010 and 2016 was 8.39% in the UK, compared with 8.36% in Germany and 8.27% in the USA. On these numbers, the UK is indeed the highest, albeit by a margin in the second decimal place: under a billion pounds separates the UK and Germany. (This very slim margin appears larger in the FT chart due to growth rates being annualised and then rounded to 1 decimal place, implying UK growth of 8.7% versus German growth of 8.1%, a difference of 0.6 percentage points rather than the actual difference in the IMF data of 0.03 percentage points.)

However, according to the revised figures, real per capita growth in the UK over this period was only 7.7%: total nominal GDP growth between 2010 and 2016 was revised down by around one percentage point in the 2022 data, culminating in lower cash GDP of around £17 billion by 2016.  Smaller adjustments to inflation estimates mean that real GDP growth was revised down by around 0.7 percentage points, from 13.4% to 12.7%. Alongside unchanged population estimates, the result is that official real GDP per capita was revised down by around £340 (in 2019 prices) by 2016 – an amount approximately equal to a third of the average household energy bill in that year.

Chart showing downward revisions to UK nominal GDP growth between 2011 and 2016

These revisions are summarised by the ONS here, and their sources are discussed here. The bulk of the revisions are due to the contribution of the insurance industry to GDP being revised down by the use of Solvency II regulatory data, as well as improvements to the way pension schemes are measured. In addition, and of particular relevance for the current exercise, part of the revisions are due to the ONS, “bringing through a package of sources and methods changes that improve the international comparability of the UK gross domestic product (GDP) estimates.” 

These revisions make a material difference to UK GDP, as well as its international ranking. On the basis of the latest official figures taken directly from national statistical agencies, real UK per capita growth of 7.7% during the austerity period compares with 8.4% for Germany and 8.2% for the US.

Chart comparing growth rates in US, UK and Germany between 2010 and 2016.

So, based on the most recent data, the UK did not have the fastest growth in GDP per capita between 2010 and 2016. 

Aside from this, as others have noted, focusing narrowly on the 2010-2016 period is potentially misleading. When austerity was implemented, the UK was in the process of recovering from the 2008 recession. It is likely that there was substantial spare capacity which, under strong demand conditions, could have been quickly reabsorbed into economic activity. If we start our comparison at the pre-crisis peak (2007 for the UK and US, 2008 for Germany), rather than 2010, the divergence is much greater: by 2016, real UK GDP per capita had increased by 2.8% on its pre-crisis level, compared with 5.5% for the US and 7.1% for Germany. Much of UK growth during between 2010 and 2016 was recovering losses from the recession: GDP per capita did not reach pre-2008 levels until 2014, compared to 2011 for Germany and 2012 for the US.

As Chris Giles notes, “Most economists now accept that the sharp reductions in public spending between 2010 and 2015 delayed the recovery from the financial crisis”. Comparing outcomes with pre-crisis levels is not, therefore, “baseline bingo” as claimed by Rupert Harrison. These outcomes are hard to square with Harrison’s claim that this is “what catch up looks like”.

Chart showing real GDP per capita between 2007 and 2016 in US, UK and Germany

These data revisions highlight the dangers in drawing strong conclusions – particularly about politically loaded topics – from small differences in data that are subject to measurement error and revision. It is inevitable that an FT article claiming that UK growth per head was highest in the G7 during the main austerity years will be used as justification for austerity policies. But, on the basis of the most recent and accurate data available, the claim is false. UK GDP growth was relatively strong by international standards (and may yet be revised back to the top of the table) but this statement ought to be placed in its proper context, using a variety of data sources and an understanding of their strengths and weaknesses.

Nominal GDP (YBHA)Real GDP (ABMI)
Year2021 Blue Book2022 Blue Book2021 Blue Book2022 Blue BookIMF 2022 WEO
20101,612,1951,612,3811,884,5151,876,0581,884,515
20111,669,5091,664,2111,911,9831,896,0871,911,983
20121,721,3551,713,2411,940,0871,923,5511,940,087
20131,793,1551,782,2961,976,7551,958,5571,976,755
20141,876,1621,862,8272,035,8832,021,2252,035,883
20151,935,2121,920,9982,089,2762,069,5952,089,276
20162,016,6381,999,4612,136,5662,114,4062,136,566

Data are in millions of pounds (2019 pounds for the real data). Data downloaded from ONS and IMF websites on 20th March 2023. Note that the 2022 Blue Book dataset was only published on the 31st October 2022, too late for inclusion in the IMF’s October 2022 World Economic Outlook. The revisions were initially introduced (and reported on) in August 2022, the quarter before the Blue Book publication.

Advertisement

“A remarkable national effort”: the dismal arithmetic of austerity

Rob Calvert Jump and Jo Michell

In a recent tweet, George Osborne celebrated the fact that the UK now has a surplus on the government’s current budget. Osborne cited an FT article noting that “… deficit reduction has come at the cost of an unprecedented squeeze in public spending. That squeeze is now showing up in higher waiting times in hospitals for emergency treatment, worse performance measures in prisons, severe cuts in many local authorities and lower satisfaction ratings for GP services.”

It is a measure of how far the debate has departed from reality that widespread degradation of essential public services can be regarded as cause for celebration.

The official objective of fiscal austerity was to put the public finances back on a sustainable path. According to this narrative, government borrowing was out of control as a result of the profligacy of the Labour government. Without a rapid change of policy, the UK faced a fiscal crisis caused by bond investors taking fright and interest rates rising to unsustainable levels.

Is this plausible? To answer, we present alternative scenarios in which actual and projected austerity is significantly reduced and examine the resulting outcomes for national debt.

Public sector net debt (the headline government debt figure) in any year is equal to the debt at the end of the previous year plus the deficit plus adjustments,

jump-deficit-eqn

where PSND  is the public sector net debt at the end of financial year, PSNB is total public sector borrowing (the deficit) over the same year, and ADJ is any non-borrowing adjustment. This adjustment can be inferred from the OBR’s figures for both actual data and projections. In our simulations, we simply take the OBR adjustment figures as constants. Given an assumption about the nominal size of the deficit in each future year, we can then calculate the implied size of the debt over the projection period.

What matters is not the size of the debt in money terms, but as a share of GDP. We therefore also need to know nominal GDP for each future year in our simulations. This is less straightforward because nominal GDP is affected by government spending and taxation. Estimates of the magnitude of this effect – known as the fiscal multiplier – vary significantly. The OBR, for instance, assumes a value of 1.1 for the effect of current government spending.  In order to avoid debate on the correct size of the nominal multiplier, we assume it is equal to zero.[1] This is a very conservative estimate and, like the OBR, we believe the correct value is greater than one. The advantage of this approach is that we can use OBR projections for nominal GDP in our simulations without adjustment.

We simulate three alternative scenarios in which the pace of actual and predicted deficit reduction is slowed by a third, a half and two thirds respectively.[2] The evolution of the public debt-to-GDP ratio in each scenario is shown below, alongside actual figures and current OBR projections based on government plans.

jump-deficit2

jump-deficit

Despite the fact that the deficit is substantially higher in our alternative scenarios, there is little substantive variation in the implied time paths for debt-to-GDP ratios.  In our scenarios, the point at which the debt-to-GDP ratio reaches a peak is delayed by around two years. If the speed of deficit reduction is halved, public debt peaks at around 97% of GDP in 2019-20, compared to the OBR’s projected peak of 86% in the current fiscal year. Given the assumption of zero nominal multipliers, these projections are almost certainly too high: relaxing austerity would have led to higher growth and lower debt-to-GDP ratios.

Now consider the difference in spending.

Halving the speed of deficit reduction would have meant around £10 billion in extra spending in 2011-12, £8 billion in 2012-13, £19 billion in 2013-14, £21 billion in 2014-15, £29 billion extra in 2015-16, and £37 billion extra in 2016-17.  To put these figures into context, £37 billion is around 30% of total health expenditure in 2016-17.  The bedroom tax, on the other hand, was initially estimated to save less than £500 million per year.  These are large sums of money which would have made a material difference to public expenditure.

Would this extra spending have led to a fiscal crisis, as supporters of austerity argue? It is hard to see how a plausible argument can be made that a crisis is substantially more likely with a debt-to-GDP ratio of 97% than of 86%. Several comparable countries maintain higher debt ratios without any hint of funding problems: in 2017, the US figure was around 108%, the Belgian figure around 104%, and the French figure around 97%.

It is now beyond reasonable doubt that austerity led to increases in mortality rates – government cuts caused otherwise avoidable deaths. These could have been avoided without any substantial effect on the debt-to-GDP ratio. The argument that cuts were needed to avoid a fiscal crisis cannot be sustained.

 

[1] There is surprisingly little research on the size of nominal multipliers – most work focuses on real (i.e. inflation adjusted) multipliers.

[2] We calculate the actual (past years) or projected (future years) percentage change in the nominal deficit from the OBR figures and reduce this by a third, a half and two thirds respectively. The table below provides details of the middle projection where the pace of nominal deficit reduction is reduced by half.

jump-deficit-table

2015: Private Debt and the UK Housing Market

This report is taken from the EREP’s Review of the UK Economy in 2015.

In his 2015 Autumn Statement, Chancellor George Osborne gave a bravura performance. He congratulated himself on record growth and employment, falling public debt, surging business investment and a narrowing trade deficit. He announced projections of continuous growth and falling public debt over the next parliament.

While much of this was a straightforward misrepresentation of the facts – capital investment has yet to recover from the 2008 crisis and the current account deficit continues to widen – other sound bites came courtesy of the Office for Budget Responsibility. The OBR delivered the Chancellor an early Christmas present in the form of a set of revised projections showing better-than-expected public finances over the next five years.

When, previously, the OBR inconveniently delivered negative revisions, the Chancellor responded by pushing back the date he claims he will achieve a budget surplus. In response to the OBR’s gift, however, he chose instead to spend the windfall.  This is a risky strategy because any negative shock to the economy means he will miss his current fiscal targets – targets he has already missed repeatedly since coming to office.

As it turns out, these negative shocks have materialised rather quickly. Since the Chancellor made his statement a month ago, UK GDP growth has been revised down, the trade deficit has widened and estimates of borrowing for the current year have increased.

ca-forecasts

In reality, the OBR projections never looked plausible. The UK’s current account deficit – the amount borrowed each year from the rest of the world – is at an all- time high of around 5% of GDP. Every six months for the last three years, the OBR forecast that the deficit would start to close within a year; every time they were proved wrong (see figure above).  Their current assertion – that the trend will be broken in 2016 and the deficit will steadily narrow to around 2% of GDP in 2020 – must be taken with a large pinch of salt.

The current account deficit measures the combined overseas borrowing of the UK public and private sectors. In the unlikely event that George Osborne was to achieve his stated aim of a budget surplus, the whole of this foreign borrowing would be accounted for by the private sector. This is exactly what the OBR is projecting. Specifically, they predict that the household sector will run a deficit of around 2% per year for the next five years. They note that “this persistent and relatively large household deficit would be unprecedented”.

This projection has been the basis of recent stories in the press which have declared that the Chancellor has set the economy on a path to almost-certain financial meltdown within the current parliament. This is too simplistic an analysis. Financial imbalances can persist for a long time. The last UK financial crisis originated not in the UK lending markets but in UK banks’ exposure to overseas lending.

But the Chancellor’s strategy entails serious financial risks. Even though there is no real chance of achieving a surplus by 2020, further cuts to government spending will squeeze spending out of the economy, placing ever more of the burden on household consumption spending to maintain growth.

The figure below shows the annual growth in lending to households. While total credit growth remains subdued, unsecured lending has, in the words of Andy Haldane, chief economist at the Bank of England, been “picking up at a rate of knots”.

debt-growth

Moderate growth in the mortgage market may conceal deeper problems: household debt-to-income ratios have fallen since the crisis but, at around 140% of GDP, remain high both in historical terms and compared to other advanced nations. The majority of new mortgage lending since 2008 has been extended to buy-to-let landlords. These speculative buyers now face the prospect of rising interest rates and tax changes that will take a large chunk out of their property income. Many non-buy-to-let borrowers are badly exposed: a sixth of mortgage debt is held by those who have less than £200 a month left after spending on essentials.

The Financial Policy Committee has noted that these trends “… could pose direct risks to the resilience of the UK banking system, and indirect risks via its impact on economic stability”.

What is often left out of the more apocalyptic visions of a coming credit meltdown is that underlying all this is an unprecedented housing crisis in which an entire generation are locked out of home ownership. Instead of tackling this crisis, Osborne is using the housing market as a casino in the hope of keeping economic growth on track during another five years of austerity. It is a high-risk strategy. His luck may soon run out.

The report’s authors include:

John Weeks on fiscal policy

Ann Pettifor on monetary policy

Richard Murphy on taxation

Özlem Onaran on inequality and wage stagnation

Jeremy Smith on labour productivity

Andrew Simms on climate change and energy

Jo Michell on private debt

The full report is can be downloaded here.

Information on EREP is available here.

Happy Christmas from the Office of Budget Responsibility

Image reproduced from here

The sectoral balances approach to economic forecasting has come under scrutiny recently. It is certainly the case that when used carelessly, projections based on accounting identities have the potential to be either meaningless or misleading. This will be the case if accounting identities are mistakenly taken to imply causal relationships, if projections are presented without a clear statement of the assumptions about what drives the system or if changes taking place in ‘invisible’ variables such as the rate of growth of GDP are not identified (balances are usually presented as percentages of GDP).

Used with care, however (or luck, depending on your perspective), the approach is not without its merits – as I have argued previously. If nothing else, the impossibility of escaping from the fact that in a closed system lending must equal borrowing imposes logical restrictions on the projections that can be made about the future paths of borrowing in a ‘closed’ macroeconomic system.

Which brings us to the Chancellor’s Autumn Statement and the OBR’s rather helpful projections. As Duncan Weldon notes, the OBR are likely to receive a rather warmly written card from the Chancellor’s office this Christmas. While true that the OBR have, in the past, been less than helpful to the Chancellor, one can’t help but wonder about the justification for announcing the OBR projections at the same time as the Chancellors’ statements. Why are the OBR projections not made known to the public at the same time that they are made available to the Chancellor?

But back to sectoral balances. The model used by the OBR produces projections which comply with sectoral balance accounting identities. Four are used: those of the public sector, the household sector, the corporate sector and the rest of the world. The most closely watched is of course the public sector balance. The headline result of the OBR forecasts is that the public sector will run a surplus by 2019. What has so far received less attention (at least since Frances Coppola examined the projections from the March 2015 OBR forecasts) is the implication of this for the other three balances. The most recent OBR projections are shown below.

Fig-1-November-2015

Since the government is projected to run a small surplus from mid-2019, the other three sectors must collectively run a deficit of equal size. The OBR projects that the current account deficit will fall from its current level of around five per cent of GDP to around two per cent of GDP. The UK private sector must be in deficit. Interesting details lie in both the distribution of this deficit between the household and corporate sectors, and in the changes in figures since the last OBR reports in March and July.

In order to show how the numbers have changed since the previous forecasts, I have collected the data series from all three releases into individual charts.

The OBR series from these three releases for the public sector financial balance are shown below. Other than postponing the date at which the government achieves a surplus (and some revisions to the historical data) there is little difference between the three releases.

Fig-2-Public

Changes to the projections for the current account deficit are more significant. The latest projections include improvements in the projected deficit of between 0.5% and 1% of GDP, compared with the July predictions. With the current account deficit at record levels in excess of 5% of GDP, I think it is fair to say the projections look optimistic. I note that in each of the three OBR series, the deficit starts to close in the first projected quarter. Put another way, the inflection point has been postponed three times out of three.

Fig-3-ROW

Things start to get interesting when we turn to the corporate sector. Here the projections have changed rather more significantly. Whereas the previous two data series showed the corporate sector reversing its decade-long surplus in 2014 and finally returning to where many think the corporate sector should be – borrowing to invest – the new series contains significant revisions to the historical data. As it turns out, the corporate sector has remained in surplus, lending one per cent of GDP in Q2 2015. The corporate sector is not now projected to return to deficit until Q3 2018.

Fig-4-Corporate

Since the net financial balance for any sector is the difference between ex post saving – profits in the case of the corporate sector – and investment, these revisions imply either falling corporate investment, rising profits, or both.

The data series for corporate investment are shown below. The historical data have been revised down significantly. Investment in Q2 2015 is 1% of GDP lower than previously recorded. (This is hard to square with Osborne’s statement that ‘business investment has grown more than twice as fast as consumption’.) The reduction compared to previous forecasts widens in the projection out to 2020. Nonetheless, it is hard to escape the conclusion that the projections are extremely optimistic. By 2020, business investment is expected to reach twelve per cent of GDP, higher than any year back to 1980.

Fig-5.Investment

What of business profits? These are shown in the table below, taken from the OBR report. It seems that corporate profit grew at 10% year-on-year in 2014-15, despite GDP growth of around 2.5%. While projected growth rates decline, corporate profit is expected to grow at over 4% annually in every year of the projection out to 2021 (in a context of steady 2.5% GDP growth). There is not much sign of Goodhart–Nangle in these projections.

Fig-6-Profits

So, to recap: by 2020 we have government running a surplus just under 1% of GDP, a current account deficit of 2% of GDP and a corporate sector deficit around 1% of GDP. Those with a facility for mental arithmetic will have already arrived at the punchline – the household sector will be running a deficit of around 2% of GDP. In fact, given data revisions, the household sector appears to be already running a deficit close to 2% of GDP – a deficit which is projected to remain until 2021 (see figure below).

Fig-7-HHAs a comparison, note that in the period preceding the 2008 crisis, the household sector ran a deficit of not much over 1% of GDP, and for a shorter period than currently projected.

The OBR has this to say on its projections:

Recent data revisions have increased the size of the household deficit in 2014 and we expect little change in the household net position over the forecast period, with gradual increases in household saving offset by ongoing growth of household investment. Available historical data suggest that this persistent and relatively large household deficit would be unprecedented. This may be consistent with the unprecedented scale of the ongoing fiscal consolidation and market expectations for monetary policy to remain extremely accommodative over the next five years, but it also illustrates how the adjustment to fiscal consolidation assumed in our central forecast is subject to considerable uncertainty.  (p. 81)

Perhaps there is something to the sectoral balances approach approach after all. One can only wonder what Godley would make of all this.

Jo Michell