Accounting for national success and failure: Rethinking the UK case

Abstract This article presents some basic political arithmetic on UK economic performance, including empirics on the sources of new job creation and regional differences. These empirics support an argument about the need for new measures and concepts of national success and failure. This is so because, as we show in the UK case, the standard post 1940 economic measures of GDP and unemployment give a seriously misleading picture of national success. This is an opportunity for accountants to join with others in devising new measures and concepts.


Introduction
We are currently seeing a revival of interest in the national economy in a world which until recently saw globalisation and regional integration within Europe as established and irreversible. History does not repeat itself but the present form of early 21st century globalisation is coming under pressure in the early 2010s much as the pre-1914 form of internationalisation did in the 1920s: accumulating economic imbalances, changed politics within and between major powers, fragile long chains of debt and trade threaten economic crisis and disintegration. Once again, the stability of the international order requires economic adjustments which are beyond the political capacity of the major governmental players and supranational authority, as we see in the case of the on-going euro zone crisis or American down grading, or the Chinese trade surplus. This raises the question about whether and how international and regional troubles will once again lead to a revaluation and rediscovery of the national economy in the 2010s as they did in the 1930s. We do not yet have a 1930s model of national prosperity through autarchy but there are, for example in the UK, growing doubts about whether the national interest is being served by the current pattern of trade and specialisation. Hence after 2008 the British political classes increasingly talk about the need for a 'rebalancing of the economy' which means more domestic manufacturing and less dependence on financial services (Froud, Johal, Law, Leaver, & Williams, 2011, pp. 4-12). More recently, opinion has shifted, partly in consequence of the UK government's award of the Thameslink carriage building contract to Siemens that shifted assembly to Germany and the impending redundancies at the Derby train building factory of Bombardier. The result has been media angst and the redefinition of public procurement as an industrial policy issue (e.g. Daily Mail, 16th July 2011; Financial Times, 18th July 2011; Guardian, 11th July 2011; Observer, 17th July 2011).
There are many ways of approaching this rediscovery and revaluation of the national economy and this article focuses on issues around the conceptualisation and measurement of national success and failure by presenting empirics and argument about the British case. The article which does this is organised in a relatively straightforward way into four sections plus a conclusion. The first section deals with how the standard economic measures of Gross Domestic Product (GDP) and unemployment gave false readings of national success before 2008 and did not register unsustainability in the UK. The second section considers the possibility of an alternative business model approach and how this is adumbrated in earlier 1970s discussions of deindustrialisation. Sections three and four then explore different aspects of how standard job creation and unemployment measures obscure key aspects of the UK's current national problems. In particular, they do not register the dependence of much new private sector employment on public funding which directly sustains the state sector; nor do unemployment rates capture the broader problem of surplus population and dependence on benefits in the ex-industrial areas of the North and West. A short conclusion suggests that the irrelevance of standard economic measures is an opportunity for socially minded accountants and others to devise new and more relevant measures and concepts. The discussion of the British case is of broader interest because it raises issues about national success and failure which are relevant to other high income capitalist countries; and also about whether and how the national economy is the relevant unit of analysis.

Post 1940 economic measures of success: GDP and unemployment
If methods format the world, so too do technical measures. But, in considering economic measurement, we should remember that the history of economics (and its measures) is part of a larger history which is as much cultural as technical. This is the point which is ably made in Esty's (2004) cultural history of England between the wars. In the title of Esty's book, England is not defined as a sinking island but as a 'shrinking island' because the decline of Empire leads to a discovery of insular identity and national culture. The intellectual projection of national identity then figures not only in the essays, poems and novels of canonical literary figures like T. S. Elliot or Virginia Woolf, but also in the economic writings of J. M. Keynes. This is hardly surprising because Keynes was, as Esty (2004, p. 165) observes, a 'Bloomsbury intimate' with a cultural hinterland nicely brought out in Skidelsky's three volume biography (2005). And thus Keynes' General Theory of 1936 combines two key elements of old and new. The older element is the scientific ambition to construct an explicit 'general theory' of capitalist economic dynamics which rested on the heroic assumption that the 'marginal propensity to consume' is always less than one and that capitalism therefore has a tendency to under consume because investment does not steadily compensate in a world where 'animal spirits' meet Knightian uncertainty. The newer element is an implicit (and taken for granted) national frame: it is the national stock exchange and ignorant mass investors who amplify cyclical trends; and the national government and its central bank are the only managers capable of delivering the necessary palliative which is a 'somewhat comprehensive socialisation of investment' around low interest rates and modest expectations of return, In the ensuing 20 years, the prospect of this kind of radical Keynesian intervention to stabilise investment was turned into an anodyne practice of demand management and Keynesianism; just as Keynesian theory was domesticated by mainstream economics as a special case of sticky wages (Leijonhufvud, 1968). But, from our point of view, the most important point is that any and all activist national strategies for economic management depended on new concepts and measures which (after a lag) were later developed in just a few years in the early and mid-1940s. National levels of economic activity and welfare were encapsulated into two standard outcome metrics of national income or output (as in gross domestic product) and of the rate of unemployment; the corollary measures of dynamic national success were growth of GDP and job creation. These privileged measures combined bricolage and system building. The bricolage came through the incorporation into economics of existing administrative measures of unemployment which were a knowledge by product of techniques of social insurance, which tied the visible extent of unemployment to variable national rules. These changed when social insurance, introduced in Britain after 1911, was extended under pressure of structural inter-war mass employment, and then changed again when social insurance was consolidated by the post Beveridgean settlement. The element of system building was provided by new kinds of national income accounting. The pioneering inter-war efforts of Clark and Kuznets were, under pressure of British war time circumstance, developed into measures of national income. These measures were the prerequisite for economic technologies to manage inflationary aggregate demand, first introduced with Kingsley Wood's 1941 budget and its supporting White Paper (Cmd. 6397, 1942). This prosaically titled official text, An Analysis of the Sources of War Finance and an Estimate of the National. Income and Expenditure in 1938, 1940and 1941, was the first to be constructed in and through the new measurement system.
When the first official national accounts were published in the United States in 1947, the US and other high income countries entered a post-war world where trajectories of national success and failure were understood through the new 'macro-economic' measures of GDP growth, unemployment rates and job creation. This was how everybody recognised they were living through the post-war long boom, which the French called 'les trente glorieuses' (Fourastié, 1979), even as economists differed about its pre-conditions and ending. And although, much changed after 1979, the dominant national frame was carried over. Thus, when a new post-1980s golden age was discovered by Stock and Watson (2002) and endorsed by Bernanke (2004), 'the great moderation' was practically defined as a moderation in the amplitude of fluctuations in quarterly American GDP growth rates. The two key measures were also used as the benchmark of relative national success. The superiority of the American model in the 1990s was upheld on the grounds that it generated growth and jobs which the German and Japanese models did not (even though they produced manufactured exports). New Labour's economic success and Chancellor Brown's repeated claims in the 2000s about 'the end of boom and bust' could not be doubted because they were statistically corroborated by higher rates of GDP growth and lower rates of unemployment. Two decades of certitude about how the standard measures captured achievement are epitomised by the vain glory of Gordon Brown in his 2005 Budget Statement which opened with a claim that: 'Britain is today experiencing the longest period of sustained economic growth since records began. . . Today I can report economic growth for the fiftieth consecutive quarter. . . This year and next the euro area is forecast to grow at 1.5% and then just over 2 per cent, Japan even more slowly at less than 1 per cent. . . And our forecast for British growth this year is 3 to 3.5 per cent and in 2006 2.5 to 3 per cent. So it is Britain and North America that have over the last eight years grown at twice the rate of most of our G7 competitors, our living standards also rising twice as fast. . ., . . .In other countries high unemployment remains today the dominant economic issue. If Britain today had German, French or euro levels of unemployment we would have 2 million fewer jobs and if we had America's higher level of unemployment, there would be one third of a million fewer jobs. The official figures published today show that since 1997 we have created two million one hundred thousand jobs. This week, every working week, another 125,000 men and women are finding new jobs. . . ' Brown (2005).
The fundamental post-Lehman question about 'why did nobody see it coming?' was first posed by Queen Elizabeth when she opened the LSE's new building (Daily Mail, 6th November 2008). It was feebly answered by the two British Academy professors who argued that 'many people did foresee the crisis' but were wrong about the form, timing and ferocity of crisis (Besley & Hennessy, 2009). More seriously, we would argue that politicians, regulators and central bankers were blind partly because of models and partly because of measures. As Bezemer (2009) has argued convincingly, with detailed quotations from heterodox economists like Wynne Godey and Michael Hudson, neo-Keynesian and Minskian theorists operating with flow of funds macro models did see that we were living through a housing bubble (even if they did not understand the risks accumulating inside wholesale banking). And it was also partly a matter of measures because, as in the quotations above from Gordon Brown, mainstream economists, politicians and regulators were fixed on the two standard outcome measures which indicated success not increasing fragility and impending instability.
This was compounded by intellectual slackness and political complacency in reading the available statistical measures because mainstream attention focused on bottom line outcomes rather than the constituent moving parts such as the different elements of final demand, the composition of investment and the objects of bank lending. All these constituent elements had been part of the mind-set of the original 1940s innovators and many of the relevant magnitudes were indeed officially measured by the 1990s. Much information could have been gained before 2008 by cross referencing information from different statistical series, even without access to flow of funds models and such like.
This general point can be simply illustrated. One telling indicator of unproductive instability can be obtained by cross referencing GDP growth against the Bank of England's standard series on housing equity withdrawal (HEW) which gives a fairly conservative measure because it excludes remortgaging for home improvement and all transaction costs which are officially classified as investment costs. Both series were widely cited before 2008 but they were very seldom put together side by side. Under the Thatcher and Blair premierships, if we take nominal values, the value of HEW was larger than GDP growth under the Tories and New Labour. In the Thatcher years from 1979 to 1990, the value of HEW is £263 billion as against GDP growth of £252 billion; while, under Blair from 1997 to 2007, the value of HEW is £382 billion as against GDP growth of £376 billion. These comparisons are suggestive rather than conclusive because nobody knows how much of HEW went into final consumption demand in different periods; and how much was reinvested, for example, in buy to let property under New Labour. But the one available study of this issue by the Bank of England (2004) admits, after examination of survey data, that between 40 and 50% of (gross) housing equity withdrawal became consumption demand. This should have been the object of critical debate and, quite probably some alarm before 2007 ( Fig. 1).
Why the wilful blindness about how different indicators fitted together into an alarming pattern? That needs a larger explanation which is provided in After the Great Complacence (Engelen et al., 2011) which presents the financial crisis as an elite débâcle resulting from the fatal combination of opaque financial innovation and the hubris of politicised expertise, especially amongst central bankers like Mervyn King and Ben Bernanke who did not recognise that the bricolage of the markets was beyond control. The long run antidote is political as much as technical but must include new concepts and measures for thinking about national welfare and activity beyond GDP, as well as some recognition that the standard measures of employment can deliver seriously misleading indications.

An unsustainable national business model?
National income and output measures have been criticised by feminists, greens and other radicals on the grounds that any measure of welfare needs to include more than marketable and material goods and services. So it could be quite properly argued that, if we are interested in measuring welfare, we should impute some value to non-marketable outputs like domestic care which is crucial to social reproduction; or that we need to recognise immaterial subtleties like happiness which almost certainly do not increase pari passu with marketable output. If GDP does measure the wrong kinds of outputs, that does not explain Gordon Brown's vain glory and the widespread acceptance of his claims before 2008. The misrecognitions of the mid-2000s concern not so much the wrong bottom line on output but a failure to consider how the different elements fit together. And from this point of view, the most urgent priority is not a better measure of output and welfare levels but something like an extended concept of business model which can indicate whether and how welfare levels are sustainable. The business model concept is usually applied to profit making firms and industrial sectors as a way of thinking about the sources of cost recovery. This usage was popularised and disparaged by Lewis (1999, p. 274) in his book on the new economy where 'all it really meant was how you planned to make money'. Our approach differs from existing usage in that it explicitly focuses on two dimensions which allow exploration of public organisations like the BBC as well as private sector firms like GE or Ford (Froud, Johal, Montgomerie, & Williams, 2010). First, we would note that all firms, whether public or private have a 'cost recovery' requirement (Williams, Haslam, Johal, & Williams, 1994) whether to make the financial surplus required of public companies by the stock market or to break even in the public sector. Second, the financial requirement of cost recovery is accompanied by a need to respond to the demands and expectations of key external stakeholders. All firms, whether public or private, are embedded within social networks of obligation where key stakeholders make influential judgements about firm performance and those judgements then have important feedback repercussions on key variables such as share price in the private sector or the assessment of value for money in the public sector. In management school debates, strategists have generally discussed the extent to which established discourse anticipates the business model concept (e.g. Amit & Zott, 2001;Chesbrough & Rosenbloom, 2002;Christensen, Laegreid, & Wise, 2002). Our usage is quite different because it brings the concept into a socio-cultural field where expectations of private profit or public value for money are variable, just as the key stakeholders and what they demand will be different in the German and UK private sectors; furthermore, viability and credibility are socio culturally mediated by narratives of purpose and achievement which circulate between firm and stakeholders (Froud, Johal, Leaver, Phillips, & Williams, 2009).
How can this concept be transposed so that we could better understand a national economy which is also a democracy? Financial viability and stakeholder credibility are brought together through employment because jobs are the primary way in which welfare is distributed in high income economies with or without elaborate welfare states. In the US case we have IRS data which allows us to measure the dependence on earned income; and, if we consider the middle income quintiles (quintiles 2-4) in 2006, between 76% and 83% of their income came from wages and salaries (Froud et al., 2010). If the issue of viability and credibility is connected to employment, then we see that the obvious point of reference is not 1990s American strategy literature but 1970s British debates about how deindustrialisation (or the decline of manufacturing) produces a trade constrained national economy which was unable to generate the required quantum of employment without current account imbalance and payments crisis. Singh's (1977, p. 126) classic article effectively provides a Keynesian definition of sustainable and unsustainable national business models: if manufacturing is the major source of foreign earnings, an 'inefficient' national manufacturing sector cannot generate the exports to pay for imports with an economy at full employment; and an efficient manufacturing sector is one whose exports pay for national import requirements 'at socially acceptable levels of output, employment and the exchange rate'. What we would do is insert this kind of economist's perception into a less mechanical and more socio-cultural world. In this world, an unsustainable and failing national economy can be turned into a narrative success which also has a performative dimension (up to a point) because the narrative influences the willingness of the financial markets to lend on capital account and bridge any current account deficit. The narrative and performative elements, which Singh did not consider, do not of course entirely abolish the structure of constraints which concerned 1970s Cambridge economists. Their work was forgotten because of the windfall gains from North Sea oil and gas and as long as global financial markets was relaxed about lending capital to balance current account deficits. However, that tolerance cannot be taken for granted in the UK in 2011 when the capital markets are picking off vulnerable European economies one after another. The UK is vulnerable because its trade deficit on manufactures stands at £80 billion in 2009 (OECD.Stat Extracts accessed 24/6/2010) and there is no other source of current account earnings which can steadily compensate. The oil is running out physically and high oil prices are a double edged sword as the UK imports an increasing proportion of total consumption. Services generally have a much lower propensity to export than manufacturing. Finance does make a reasonable contribution because it does generates an export surplus which grew from £11,769 million to £32,919 million between 1999 and 2009 and went some way to covering the deficit on trade in goods which increased from £29,051 million to £81,875 million over the same period (OECD.Stat Extracts accessed 24/6/2010) ( Fig. 2).
But, even in the bubble years before 2008, there was no realistic prospect that a surplus from finance would cover the deficit on manufactures: furthermore, since 2008 we understand that the private trade surplus on finance comes at the expense of offsetting public sector liabilities for bailouts and other subventions. The medium term outlook is unfavourable. And that is so, even if we ignore the complication that further public subvention of banking may be required once again after European sovereign default; and even if we suppose the financial markets will continue to lend to Britain at reasonable rates for some time because they are distracted by worse problems elsewhere. On current trends, the medium term current account deficit will increase to the point where the markets will panic. Two senior Cambridge economists (Coutts & Rowthorn, 2010, p. 15) have recently returned to the theme of payments constraint: their model projects a deficit which increases to almost 5% of GDP which is, according to Bergsten (2002), in 'a danger zone of current account unsustainability' for industrial countries. It helps greatly that Britain did not join the euro but depreciation of the pound does no more than slow continuing balance of payments deterioration, as we can see if we look at the effects of the 25% depreciation against the euro after 2007 (Fig. 3).
Britain (with exchange rate depreciation) has the same problem as the Mediterranean European Union countries (within the euro zone), i.e. a large appetite for consuming imports from Germany and low wage Asia that cannot easily be paid for by exports, because the country has no clear role in the new and more international division of labour which can give it the capacity to generate current account export earnings on the scale necessary to pay for imports. Though little discussed, the British economy going forward is balance of payments constrained as Cambridge economists in the 1970s had feared. Thus, letting go of manufacturing has turned out badly for the UK, as is recognised in a confused way in the current political and media discussions of the desirability of 'rebalancing' which do not engage with any of the practical difficulties of rebuilding broken supply chains . In the 2010s, the outcome may be worse than anticipated in the earlier apocalyptic visions of the 1970s because the jobs growth and unemployment indicators have been giving readings which camouflage worsening structural problems.

Job creation and the dependence on state funding
One of the reasons for misplaced optimism before 2008 was that the various indicators of the number of employees, workers and jobs all suggested that the British economy had become (in the aggregate) a remarkable job creation machine. Table 1 presents a broad overview of the number of UK jobs in the aggregate and in various key sectors. The number of jobs did not steadily increase under the Conservatives (from 1979 to 1996) but there were nevertheless more than one million extra jobs in 1997 after UK employment had increased from 27.3 million to 28.6 million. Then, in the New Labour years, more than 2 million extra jobs were created so that UK employment had increased from 28.6 to 31.0 million. This outcome of jobs growth is all the more remarkable if we consider the sectoral sub-trends. Manufacturing employment declines unsteadily under the Conservatives from 6.6 to 4.2 million between 1979 and 1997 and then remorselessly every year under New Labour from 4.3 to 2.6 million between 1997 and 2009. Quite remarkably, the finance sector does nothing for increased employment in its long boom from the early 1990s because increases in turnover and assets did not require extra workers; so, the finance sector which employed just 1.0 million at the point of deregulation and 'Big Bang' in the 1980s employed no more than 1.1 million in 2009. But other services of all kinds hugely increased employment by some 8 million over the whole period from 17.3 million jobs in 1979 to 21.7 million in 1997 and then to 25.5 million in 2010, when 82.5% of the workforce was engaged in services. If Thatcher had embarked on an enterprise experiment of flexible labour markets and low rates of taxes on high incomes and corporate profits, this was apparently vindicated by the results in terms of job creation which Gordon Brown presented triumphantly.
But first impressions are misleading because the categories of official statistics conceal the increasing dependence of the economy on state-funded employment; and, insofar as the increase in employment depends on increased government expenditure, the outcome is both unsustainable in the long run and has very little to do with the Thatcher experiment in labour market flexibility and other much vaunted reforms. None of this was obvious because official statistics have traditionally classified workers and jobs as public or private sector, according to the identity of the employer as public sector organisation or privately owned firm. But the position is increasingly confused after 1979: first, by privatisation which switches workers between state and private sector; and then, by outsourcing and sub-contracting which creates an increasingly important para-state sector where private employment is sustained by public funding. All these confusions have one result, the official statistics flatter the job creating performance of the private sector and obscure the extent to which new job creation since Thatcher has depended on government spending. In the case of privatisation, the result is a kind of social book-keeping adjustment in the employment figures because the public sector workforce diminishes and the private sector workforce increases with each privatisation. The overall effect can be roughly measured by counting the workforce at the point of privatisation as we have done in 16 major cases. 1 The calculation shows that in the Thatcher and Major years from 1979 to 1997, privatisation transferred some 825,000 individual public sector workers into the private sector: strikingly, the number of private workers would actually have fallen without this adjustment. It is altogether more difficult to work out the effects of outsourcing and sub-contract which grows the para-state sector, where an increasing number of employees (in activities like nursery education or care for the elderly) work for a private firm which is publicly funded. We have developed two methods for estimating the dependence of job creation on public funding: first, we can construct a long-run series for the whole of the post-1979 period by using the aggregate total of education, health and public administration jobs as a three sector proxy for publicly funded employment; second, for the post-1997 period we can build up a more fine grained, multi sector analysis of the number employed in the para-state by adding together sectoral estimates based on the importance of government demand in different sectors. The simplest way to construct a long run series of publicly funded employment is by approximation if we treat all employment in health, education and public administration as publicly funded. This approximation captures the truth that public funding sustains almost all the employment in these sectors. The three sectors are variously supported by direct central state expenditure on objects like schools and hospitals; local and national government subvention of charges by private care homes and nursery schools; and tax breaks and subsidised training which benefit private medicine and education. These three sectors are also the large scale employers of state and para-state workforces; total employment in the three sectors increases from 5.7 million to 8.5 million between 1979 and 2010. If these sectors are not 100% dependent on state funding, The result is not perfect but it provides a good, easily calculated, long term proxy and the results are quite startling. Over the whole period 1979-2010, the three sectors account for 2.84 million extra employees or 67.5% of the total increase of 4.3 million employees. And this is not a recent development driven by New Labour profligacy which encouraged an enlarged state that the Tories did not. Under Thatcher and Major from 1979 to 1997, employment in the three sectors increased by one million and this accounted for 86% of the total 1.2 million increase in employees. Mrs. Thatcher's practice of government was complicated and interesting because her public rhetoric and her theatre of reform was pro-market and private sector; but, in an undisclosed way through government expenditure, she built the state. And New Labour does no more than continue this practice, so that in the Blair years from 1997 to 2010 the three sectors account for 1.9 million extra employees, equivalent to some 60% of the total increase in employees. The three sectors actually account for a smaller per cent of the total increase in employment under New Labour, although the absolute increase in employment in these three sectors each year is larger under New Labour with average increases of 132,000 per year, as compared with 60,000 per year under Thatcher and Major. The most important point is that three sector employment increases virtually every year 2 by an average of 92,000 thousand from 1979 to 2010.
It is possible to construct a more fine grained analysis of developments under New Labour using reworked official statistics to discover which sectors generated the extra jobs. This is necessary because state employees are now working alongside increasing numbers of publicly funded but privately employed workers in social care, nursery education and outsourced local government services. In this case we proceed by estimating publicly funded employment for 37 sectors. This is done by reworking the official Standard Industrial Classification (SIC) 4-digit activity group totals of employees in the ONS' Annual Business Inquiry (ABI), which has been published every year since 1998. Publicly funded employment is calculated by multiplying total employment in each sector in 1998 and 2008 by an estimate of the proportion of final demand in that sector attributable to public spending. The procedure is robust because employment is concentrated in a few sectors (health, education, social work and social control) and it discloses 1.7 million publicly funded and privately employed workers. Effectively, the three sector long-run proxy and the multi sector estimate overlap so that the one corroborates the other (Fig. 4). The multi sector estimates for the New Labour years confirm that state and para-state (S&PS) employment makes a major contribution to job creation. As Fig. 5 shows, S&PS employment increased by nearly 1.3 million from 6.2 to 7.5 million between 1998 and 2008. This accounted for no less than 55% of the total increase of 2.3 million in the number of employees from 24.4 million to 26.7 million on the ABI measure. The pattern from 1998 to 2008 is one of sustained increase on a large base with a 17% increase in S&PS employment over the decade after 1998, so that S&PS together employ nearly 7.6 million or 28% of the workforce by 2008. The weight and force of S&PS employment creation, as well as the huge base, is such that, if the UK has a 'leading sector', it is the state. And this represents a long run continuity which aligns the employment outcome of the 1990s and 2000s with pre-1979 trends. The difference is that authors like Bacon and Eltis (1976) could then argue that the expansion of state employment was 'crowding out' the private sector but 30 years later the long run trend now looks more like filling in for an anaemic private sector.
In terms of gender and contribution to the creation of female employment, S&PS made a much larger contribution. Employment in the S&Ps sector is heavily gendered because rank and file workers in health and education are disproportionately female. In both 1998 and 2008, just over 69.3% of the S&PS workforce is female and this group splits more or less equally into half full-time and half part-time female workers, whereas only 21.2% of male S&PS workers are part-time. The end result is that the S&PS sector is dominant in the creation of new full-time and part-time jobs for women. As Fig. 5 shows over the period 1998-2008, S&PS accounts for an extra 944,027 female jobs which split 55-45 between full-time and part-time; and these new jobs account for no less than 81.6% of the total 1.2 million increase in female employment over these years. If high income capitalist countries are changing because wage earning households are increasingly dependent on two wage earners, in the UK case, the S&PS sector more than any other puts the second wage earner into the average household.
The message of this section is twofold. First, we should and can develop new measures of employment creation which focus on the drivers of job creation which are as relevant as the number of jobs created. Second, that any new measures will raise questions about UK dependence on state funded employment and the unsolved problem of the weakness of the private sector. The 30-year enterprise experiment cannot have succeeded if the creation of extra jobs was, under Tories and Labour, dependent on (unsustainable) continuous increases in public spending on health and education. And, as we will argue in the next section, cut backs in public expenditure will now greatly aggravate the regional problem in the UK.

Regional problems and the end of the national settlement
If the British are in the middle of a process of rediscovering the value of the national economy, there are many awkward and difficult issues which they have yet to take on board. The national economy cannot now be a kind of "found object" as it was for Keynes in the 1930s because regional differences are sharply increasing in the 2010s. The paradox is that the national economy is now being rediscovered as it is ceasing to exist in its traditional political and economic sense as a political space of negotiation and an economic space of redistribution which guarantees minimum standards for all regions and individuals within its boundaries. If that claim appears to be hyperbole, that is because the standard outcome measures on unemployment produce indicators obscure the pre-2008 position of the ex-industrial areas which had lost manufacturing employment and gained nothing except para-state jobs. Unemployment rates were relatively low, even in the ex-industrial regions; but that was because other forms of benefit like invalidity benefit were being used to maintain a surplus workforce. This outcome was in itself a perverse result of the privileging of unemployment rates as a politically sensitive indicator. Humane family doctors and realistic benefits officers effectively colluded to park up workers on invalidity benefit when there were no suitable jobs within or outside their locality. In these circumstances, the relevant measure of dependence and discouragement is not the percentage of the workforce unemployed but the share of the employable population on all forms of benefit, and that is much higher. Once again, a very different non-standard story can be told by putting together fragments of evidence and, in this case, simply adding up the numbers of those on different kinds of benefit. In Table 2 we have calculated the number and proportion of the economically active population who are out of work and drawing some form of benefit on grounds of unemployment or invalidity. As a point of reference we also include a calculation of benefits against all aged 16-64, where the denominator includes the discouraged and those ineligible for benefit. The regional contrast is quite startling and the predicament of the ex-industrial regions is dire because here the settlement after deindustrialisation created a new Speenhamland which offered full maintenance for a wholly unemployed industrial population. The North East and Wales have over 20% of their economically active population on out of work benefits and the West Midlands and Scotland around 17.5%. By way of contrast, the percentage on some form of out of work benefit in the South East is no more than 9.0%. London has a much higher rate of 16.1% on out of work benefit, indicating the extent of the deprivation which co-exists with great wealth in that city.
If part of the problem is the ex-industrial areas which have no autonomous private sector capable of job creation, the other part of the problem is that London (or London and some select parts of the South and East) are drawing away from the rest of the economy because they do have job creating private sectors (even if finance creates no new net jobs). In the previous section, we presented national figures using the Centre for Research in Socio-Cultural Change (CRESC) method for calculating publicly funded employment in the (private) para-state sector and then added these para-state totals to those of state employees (Buchanan et al., 2009, pp. 16-19). Table 3 presents regional figures using the same method. The story is that employment in finance may be flat but London was creating jobs (and not only para-state jobs) unlike any other region. Between 1998 and 2007, London created more than 400,000 jobs and more than two-thirds of these jobs were in the private sector and not dependent on public funding. This was significantly higher than the next best performing region, the South East, which managed to create 333,000 extra jobs, of which 56% were private sector; and hugely different from the worst performing region of the West Midlands where private sector employment was declining and all of the 65,000 job net increase and more was publicly funded.
After drawing this regional contrast in the New Labour years, we can make two important supplementary points. First, the extra jobs in London were of no benefit to native born British because there is no net in-migration into London from other British regions in the 2000s (as there was from the North and the West into the Midlands and South East in the 1930s). A report by Gordon, Travers, and Whitehead (2007) notes that employment in London increased by some 388,000 between 1997 and 2006; however, more than 85% of the new jobs are held by London residents born outside the UK. The authors broaden the analysis to consider all net new jobs at London workplaces, including those who travel into London to work and find that all of the new employees were born outside the UK. Second, the contrast between London and the ex-industrial  regions is even more marked after 2008 because the job losses of recession are concentrated in the North and West. In the worst economic downturn since the war, between 2007 and 2010, some 712,500 jobs were lost nationally but more than 85% of that total or some 621,200 were lost in the ex-industrial regions of the West and North. By way of contrast, the East, East Midlands and South East regions each suffered job losses of no more than 50,000; and the number of jobs in London actually increased by just over 5000. The huge difference between patterns of job creation is such that the second standard indicator of output already shows a widening gap between London and the regions. Fig. 6 compares gross value added (GVA) per capita 3 in the UK regions against London whose income serves as a base of 100 in both 1989 and 2009. The point of the comparison is that all regions (including the South East) fall behind London over these 20 years but the fall is greatest in the West and the North. Thus the three laggard regions of the North East, West Midlands and Wales fall by an average of 9 percentage points against London GVA per capita, so that at the end of this period each one of these regions has a GVA per capita which is less than half that of London. The Welsh GVA per capita was 54% of that in London in 1989 and only 43% by 2009: if this trend were to continue, Welsh GVA by 2029 would be no more than one-third of London GVA per capita.
Under New Labour, the ex-industrial regions were being kept afloat by a kind of social settlement which combined publicly funded employment and benefits: this is now being withdrawn with the impending public expenditure cuts. The evidence in this section raises questions about whether the national is still the relevant unit of analysis because the British are not all in the same boat but, increasingly, they are locked into different compartments with few of the traditional connections established by internal migration or politically sponsored redistribution. This has not been taken on board in the Keynesian enthusiasm for reflationary plan B to encourage growth. In a regionally divided economy, Keynesian reflationary expansion to expand employment and output is irrelevant to new realities: most forms of fiscal stimulus would not effectively reach the ex-industrial regions where half a million jobs have been lost; meanwhile stimulus may be unnecessary in London where employment is increasing. The immediate problem in Britain (and indeed in most high income capitalist countries) is distribution not growth.

Conclusion
This article makes an argument and presents illustrative empirics on the British case which shows how the standard economic measures of GDP and unemployment can be misleading measures of national success and failure. Or, more exactly, that these measures are misleading when they are abstracted from context and read uncritically, as they were in Britain before the financial crisis because they then assumed a specific capitalist context (with, for example an autonomous private sector) which did not exist in the British case. This basic point needs to be made clearly because all the misunderstandings from the pre-2008 period are being carried over into solutions, insofar as the standard Left Keynesian fix for recession in 2008, or double dip in 2011, is reflation to boost GDP and generate employment. As readers of this article will understand, this kind of generic preference does not engage with the structural specifics of trade constraint and private sector anaemia in the British case.
The implication is that we need to engage national specifics via new measures and a more sophisticated reading of existing measures. As this article demonstrates, this kind of shift is perfectly practical and immediately within the realm of the possible because it requires not so much a radical change of paradigm but a progressive kind of bricolage which puts together different existing series, reworks figures by estimation and thereby puts together a case specific understanding of what drives and limits employment creation in different national cases. These are good reasons for putting this kind of political arithmetic at the centre of a revitalised economics, but it is also true that the current cadre of economists (mainstream or behavioural) have invested their intellectual capital in other methods and objects so that they are probably incapable of the necessary shift.
Is this not then a huge opportunity for a new generation of socially minded accountants? After all, their core skills already include the capacity to read through company accounts and contextual information so that they see how bottom line results are produced; and many of these company accounts skills should be transposable to reading national accounts. And, when it comes to the formulation of new measures, accountants are well qualified to propose modifications and supplements. If national income accounting was devised by economists in the 1940s, the proposers of value added accounting frameworks in the 1970s, like Cox (1979) and Gray and Maunders (1980), demonstrated the role that accountants and engineers can have in devising new and socially relevant measures of welfare and distribution. If accountants can move outside their comfort zones in financial reporting and in critical commentary, they could now make a major contribution.