The UK’s poor productivity performance in the Great Recession and its aftermath: how is it to be explained?
The supply-side performance of the UK economy deteriorated in the Great Recession of 2008 and 2009, and has remained unimpressive in the hesitant recovery that has followed it. In the year to the third quarter 2007, which saw the run on Northern Rock and so heralded the financial strains of the Great Recession, national output was booming. As measured by gross value added in real terms, it rose by 4.7 per cent. From then until the third quarter of 2012 the average value of the annual change in national output was negative at minus 0.2 per cent. In other words, at the time of writing (April 2013) output remains below its level five years ago, an outcome so weak that it has no precedent in the period of modern quarterly national accounts that began in 1955. However, employment has been surprisingly resilient in the Great Recession and has recently reached a new all-time peak. As the change in output can be viewed as the sum of the change in employment and the change in output per person employed (or ‘productivity’), it is clear that the central disappointment of these years has been the stagnation of productivity. The accompanying table below shows that in the five years to autumn 2012 productivity typically fell by 0.4 per cent a year, whereas in the preceding 45 years it rose on average by about 2 ½ per cent a year. The question to be discussed here is ‘why?’.
One of the most well-known relationships in business cycle analysis is that between the rates of growth of output and productivity. Numerous studies have found that employment exhibits far greater stability than output.1 This feature of the labour market is presumably due to significant costs of hiring and firing for employers, and of job loss and search for workers.2 The relative stability of employment means that years of high growth of output tend also to be years of high growth of productivity. The first part of the paper therefore considers the extent to which the recent poor productivity figures are to be attributed to the unsatisfactory output record of the UK economy in the Great Recession period. If the setback on productivity is readily blamed on the cyclical sluggishness of output, this would be relatively encouraging for the UK’s macroeconomic outlook. It would imply that not too much has gone wrong with the long-term, underlying, structural characteristics of the economy, and that a return to higher growth of demand and output ought to be accompanied by a recovery in productivity growth. Unfortunately, a statistical exercise shows that recent productivity numbers have been significantly worse than can be explained by the past relationship between changes in output and productivity. Something fundamental, or perhaps a number of fundamental ‘somethings’, seems to have gone wrong. A review of several possible culprits for the UK economy’s supply-side reverses is then offered. One source of weakness is uncontroversial, the depletion of the UK’s oil and gas resources. Apart from that, the evidence suggests that the poor productivity figures of recent years are more likely to be attributable to mistakes in government policy than to private sector inadequacy of some sort.
Has recent productivity performance been worse than implied by the past output-productivity relationship?
As just noted, because employment tends to be more stable than output, the growth rates of output and productivity are correlated. The chart above gives the annual rates of change in national output and whole-economy output per head from 1960 and illustrates the point clearly. Given that output fell steeply in 2009 and still has not regained its earlier peak (in early 2008), it should come as no surprise that productivity growth has been beneath the average of recent decades. If the economy were behaving ‘normally’, the output-productivity correlation implies that poor output ought to be associated with poor productivity. Productivity performance can be condemned as ‘underperforming’ only if it has been worse than implied by the past output-productivity relationship.
The output-productivity relationship over the period 1960 to 2007 was therefore estimated by standard statistical methods, deriving an equation that easily met the usual significance tests.3 One way of assessing the post-2007 performance of productivity, relative to the economy’s underlying long-run behaviour, was to see whether actual performance in 2008 and later was above or beneath the values implied by the pre-2008 equation. (The actual values of output were ‘plugged into’ the equation, which was extrapolated to the third quarter of 2012.) The result was that output in 2008 and later was appreciably lower than would have been expected if the pre-2008 output-productivity relationship had survived. The chart below shows the economy’s over- and under-performance on productivity relative to the pre-2008 output/productivity relationship.4 The last few years of the Great Recession – which are those inside the box with the extrapolated outcomes – are clearly among the worst in the entire period of over five decades. (Nevertheless, it is worth saying that on this criterion of performance the five years to the third quarter of 2012 are not noticeably worse than some other multi-year periods, including some multi-year periods in the 1980s, a decade occasionally revered for its economic dynamism.)
Although future upward revisions to the official estimates of output in these years are possible, it is clear that productivity in the Great Recession has been subject to an abnormal check. In the five years to the end of 2012 the cumulative negative residual compared with the earlier relationship amounted to over 4 per cent of output. In other words, productivity growth had on average been unsatisfactory, compared with past norms, by almost 1 per cent a year for five years.
How is this shortfall to be explained? What influences might account for the weakness of the productivity numbers? Five suggestions are now offered, in no particular order. Of course, each of the candidate explanations is the more persuasive if it relates to an economic development specific to the last few years.
Explanation 1 for poor productivity: depletion of North Sea oil and gas resources
Natural resource depletion is a fact of life. Typically the highest-return oilfields and mines are exploited first, with the result that later production requires more resources of labour and capital (i.e., lower productivity) than earlier. As natural resources are depleted, the productivity of factors of production used in their extraction declines. In the case of the UK’s offshore oil and gas industry, this consideration has affected the growth rate of whole-economy productivity to some extent. Also important is that labour productivity in the oil and gas sector is exceptionally high, at least because of the heavy capital expenditure (and high capital/labour ratio) involved in this form of production. As the sector declines in relative importance, whole-economy labour productivity is reduced.
The UK’s offshore oil and gas production peaked in 1999. Since then national output excluding oil and gas has risen on average by ¼ per cent a year a more than national output as a whole. Since employment in the sector is small, the depletion of the UK’s oil and gas resources has lowered the rate of whole-economy productivity change by a similar figure, say, 0.2 per cent a year. This is part, if only a small part, of the explanation for the weakness of recent productivity outturns compared with the long-run average. But it must be noticed that this negative factor began almost 15 years ago, not at the start of the Great Recession.
Explanation 2: high levels of public expenditure
The significance of high levels of public expenditure and taxation for economic growth is much debated. Two adverse effects of a large state on productivity performance are commonly recognised. First, many economists believe that the higher are taxes, the weaker are the incentives to work and to save.5 Secondly, because most of public expenditure is paid for by taxes, it is not subject to a market test and the motivations for enterprise found in the private sector do not apply.6 Productivity growth tends therefore to be lower in the public sector than in the private sector. (Over time the result is of course that the level of productivity in state-owned entities is lower than in comparable privately- owned entities, ultimately justifying privatisation.)
As the Great Recession has been accompanied by a high ratio of public expenditure to national output, this aspect of the economic situation needs to be included in the discussion. But the subject is vast and space limited. The chart below shows that the ratio of general government expenditure to GDP has been about 45 per cent in recent years, almost 5 per cent above the 1980 – 2012 average, and well above the range of 35 – 40 per cent seen in the period of benign economic outcomes from 1992 to 2007 sometimes known as ‘the Great Moderation’. Varying estimates have been made of the damage done to a nation’s growth potential by excessive levels of taxation and public expenditure. But some negative effect, perhaps quite small at about ¼ per cent a year, seems plausible from the over-expansion of the public sector during and since the Gordon Brown premiership.
Explanation 3: the cost of renewable energy
The European Union has accepted the so-called ‘warmist’ doctrine that, because of the carbon emissions arising from modern industrialism, mankind is largely to blame for the global warming of recent decades. The EU has therefore decided that its member states must take action to curb carbon emissions, even though such action leads to the replacement of low-cost fossil fuels by high-cost renewable energy and is plainly inefficient in economic terms. As an EU member, the UK has participated in the drive towards renewable energy. The three key directives here are the 2001 Large Combustion Plant Directive, the 2003 Bio Fuel Directive and 2009 Renewables Directive. The last of these is the most significant and undoubtedly the most costly. The purpose of the 2009 Renewables Directive is, explicitly, to move towards a 20 per cent drop in the EU’s carbon emissions by raising the proportion of electricity generated by renewables (wind, wave, solar and so on) to 20 per cent by The cost of electricity generation by means of renewable energy is much higher than that by conventional methods (gas and coal firing, mostly). For example, electricity from offshore wind farms costs at least three times as much to produce as electricity from a gas-fired combined-cycle power station.
This is not the place for a lengthy discussion of the environmental impact of carbon emissions. It may or may not be proved 20 or 30 years from now that global warming has been caused predominantly by mankind. Whatever the outcome of that debate, several nations are not making major adjustments today to their policies towards energy, electricity generation and the environment. In electricity generation they continue to invest in order to minimize cost. As a result, households and companies in every country in the EU – and not just the UK – will have to pay well above the international price for electricity. Industries heavily reliant on energy usage and electricity consumption will become too high-cost compared with suppliers from other countries. They will stop investing in the UK and other EU countries. As a news story in The Sunday Times on 6th February 2011 remarked, ‘Leading chemical companies have warned the government that its energy policies will render them uncompetitive, leaving plants to “die on the vine” to quit Britain for lower-cost countries.’7
Government departments have of course had to advise ministers on the costs of the UK’s adoption of the EU’s green agenda. The Guardian has received a series of leaks from sources in the Department for Business, Enterprise and Regulatory Reform (formerly the Department of Trade and Industry) on key energy policy issues that have been and remain in dispute. Some leaked documents indicated that the cost of meeting EU targets would be between £5 billion and £11 billion a year. Indeed, according to the documents, the long-term goal of 20 per cent of total energy being from renewables would cost £22 billion.8 Because the British government has at the EU’s behest imposed methods of electricity generation that are costly and inefficient, Britain is worse-off without qualification. An unfavourable effect on labour productivity is to be expected while the switch to renewable energy is taking place. (The selection of costly methods of electricity generation may eventually prove to have been correct, in that lower carbon emissions may help ‘to save the environment’. But – as of now – that is conjecture.)
The size of the adverse effect, in terms of the effect on the annual growth rate of productivity, is for debate. In 2010 the gross value added of the ‘electricity, gas, steam and air conditioning supply’ category in the national accounts was £18.9 billion, compared with the nation’s total GVA of £1,265.0 billion. The EU-inspired drive to renewables may have reduced productivity growth by 0.1 per cent (or at most 0.2 per cent) a year while it has been (and continues to be) implemented. (Similar remarks might also be made about EU environmental legislation, with an apparent requirement for serious over-investment in water purity and hence in water supply infrastructure. Also detrimental to growth are a variety of EU directives on employment, but these have been taking effect for many years and it is unclear that the intensity of labour regulation increased from 2007.)
Explanation 4: increased financial regulation and the City of London
The UK’s financial sector has been heavily criticized as being in some sense culpable for the setbacks of the Great Recession. The popular media have highlighted the disparity between the high incomes earned by top executives in the banking industry, and the poor returns for banks’ shareholders and other stakeholders. In this context the UK’s financial sector and ‘the City of London’ are often bracketed together. However, this overlooks that the particularly high incomes earned in the City result predominantly not from domestic UK banking, but from financial service exports to (mostly) large corporate clients around the world. During the Great Moderation the boom in these exports contributed disproportionately to UK GDP growth.
Some data on these patterns is given in the accompanying table. In 1992 the UK’s exports of financial services were less than 1 per cent of GDP. Although they declined in 1992 itself, over the next 15 years they climbed at an annual compound rate of about 15 per cent. These gains reflected the application of new information technologies, which facilitated greatly increased trading volumes, and globalization, as well as a range of financial innovations. (They had relatively little to do with the booms in mortgage and other types of personal credit in UK domestic banking in the opening years of the 21st century.) In 2008 financial service exports amounted to over 3½ per cent of GDP, implying that this one part of the economy – to repeat, a mere 0.9 per cent of the economy in 1992 – contributed about ¼ per cent a year to aggregate growth during the Great Moderation.
Since 2008 the growth has stopped, as the chart below shows. The growth may merely have paused, reflecting the hostile environment of the Great Recession, or it may be attributable to the frustration of the fundamental growth drivers. The imposition of a stricter regulatory framework is an obvious barrier to resumed growth in the financial sector in the advanced countries at present. The new regulations are being applied at the national level, but they stem from international agreements (on, for example, banks’ capital/asset ratios) reached under the aegis of several supranational bodies, but most notably that of the Bank for International Settlements in Basle. In countries outside the BIS orbit the financial sector continues to grow rapidly. At any rate, as far as the UK is concerned, the loss of the dynamism in financial exports is one reason – accounting for perhaps ¼ per cent a year – that recent productivity growth has been disappointing. (Nothing is said in detail here about a further possible effect of tighter financial regulation. It is clear that, if the target rate of return on equity is given, banks must respond to an increase in the regulatory capital/assets ratio by widening loan margins. That increases the cost of finance to the corporate sector and reduces the range of capital projects that is economically viable, with an unfavourable effect on labour productivity. An effect of this sort may be a contributory factor in the UK’s current weak productivity performance, but full substantiation of the argument is not provided here.)
Explanation 5: low productivity of immigrant workers
It was noted earlier that during the Great Recession employment was surprisingly resilient, given the severity of the fall in output. However, a distinction needs to be drawn between UK-born and foreign- born workers. The years since the 2004, which saw the accession of eight East European countries to the EU, have been marked by heavy immigration into the UK of foreign people of working age. In this period about half of the immigration has been from the eight East European countries (‘the EU-8’). Despite some scare stories in the popular press about ‘benefit tourism’, most of the immigrants have come to work and have found jobs. The result during the Great Recession was a sharp dichotomy in the employment patterns of the two groups. At the end of 2007 total employment was just under 29.5 million, split between UK-born of 25.9 million (87.7 per cent of the total) and foreign- born of 3.6 million (12.3 per cent). Over the next four years UK-born employment dropped to 25.0 million (85.8 per cent), whereas foreign-born employment went up to 4.1 million (14.2 per cent). Although a case can be made that the immigrant workers took jobs away from those with long- standing UK connections and in that sense were better-qualified than the UK-born in the activities where they concentrated, there is little doubt that these activities were and remain typically low- income and low-productivity.9 Examination of 2007 – 09 employment records by an official statistician concluded, ‘EU-8 workers are predominantly employed in “Elementary occupations”.’10
So in the four-year period of 2007 – 11 the balance between UK- and foreign-born in the UK labour force changed by 2 per cent, while the average productivity of the new foreign-born workers was lower than that of the workforce as a whole. If it were assumed that the productivity of the newly employed foreign workers was two-thirds of the existing workforce, this development would explain a drop in the level of productivity of between ½ and ¾ per cent. Given that it occurred over a four- year period, the negative impact on the annual growth rate of productivity was between 0.1 and 0.2 per cent.
Conclusion: the productivity check to be blamed on government policies, not private sector failure
The discussion in this paper may not seem to have a salient unifying theme. It has been shown that the UK’s productivity performance during and immediately after the Great Recession has been mediocre, even when allowance is made for the well-established cyclical correlation between changes in output and productivity. Relative to what might be termed ‘reasonable expectations’ given the output- productivity correlation observed in the past, the shortfall has been almost 1 per cent a year. The weak numbers have been blamed here on five influences, and these influences appear miscellaneous, ad hoc and unconnected. For example, the effect of new regulatory constraints on the high-productivity cluster of City-of-London-based industries is quite separate from the impact of low-productivity immigrants on aggregate productivity. However, it could be claimed that four of the five influences reflect official policy in some way, and that only one – the depletion of the UK’s North Sea oil and gas resources – is unrelated to policy. Indeed, three of the remaining four influences (i.e., the move to renewable energy, the adoption of tighter regulatory standards in the financial system and immigration from Eastern Europe) arise to some degree from the UK’s membership of the EU. Perhaps it is not surprising that disquiet is being expressed about the cost-benefit implications of the UK’s continued participation in ‘the European construction’, with the Foreign and Commonwealth Office now conducting a review of the ‘competences’ granted to the EU.
As recognized in the discussion above, the effect of a large state sector and high taxation on economic growth is controversial. All the same the expansion of government spending in the second half of the 1997 – 2010 Labour administration was of course the result of decisions taken by politicians, not of businesspeople in the private sector. To the extent that higher taxes have hurt the economy, here too the fault lies with politicians and officialdom, not with a failure of enterprise and the private sector. The table below brings together the strands of argument in this paper, providing an apparently quite effective interpretation of the check to UK productivity in the Great Recession. A warning has to be given economists are bad at understanding the causative forces at work in economic growth and the neatness of the ‘explanation’ should not be pressed too far.
In the so-called ‘tech bubble’ of the late 1990s some commentators spoke of a New Era of much- enhanced productivity gains.11 Given the excitement and hopes generated by the revolutionary Internet-based technologies, the setbacks of the Great Recession are doubly disappointing. But the new technologies did contribute, for example, to the remarkable dynamism of the UK’s financial sector in the Great Moderation. Huge scope remains for their application in the media and publishing industries, in which the UK ought to do well because of its central position in the English-speaking world. The Great Recession has been characterized by both increased regulation and stagnant productivity. Sooner or later the arguments for deregulation will again be articulated and translated into policy. Might higher and historically more normal rates of productivity growth then be resumed?
This note expands observations originally made for the author’s column in the Economic Affairs journal, ‘Growing pains for the next government’, in 2010.