Later in the twentieth century there were repeated attempts to clear up the confusion over whether certain kinds of government spending counted as intermediate or final consumption. This was done by identifying which government activities provided non-market and free services for households (for example, schools), as opposed to intermediate services for businesses (for example, banking regulation). The distinction is not easy to make. Governments build roads. But how much of their value accrues to families going on holiday and how much to a trucking company moving essential spare parts from factory to user? Neither family nor trucking company can build the road. But the family on holiday adds to total final demand; the trucking company is an intermediate cost for businesses.
In 1982, national accountants estimated that some 3 to 4 per cent of Swedish, German and UK GDP was government expenditure that, previously categorized as final consumption expenditure, should be reclassified as (intermediate) inputs for businesses. This had the effect of lowering government’s overall value added by between 15 to 20 per cent.23 To take an example of such reclassification, in 2017 the UK telecommunications regulator, Ofcom, compelled British Telecom (a private firm) to turn its broadband network operation Openreach into a separate company following repeated complaints from customers and other broadband providers that progress in rolling out broadband around the country had been too slow and that the service was inadequate. At least part of the cost of Ofcom could be seen as beneficial to the private telecommunications sector. Yet the convention that all government spending should count as final consumption has proved remarkably resistant to change.
Now we can see why government final consumption expenditure is bigger than its value added in Figure 8. The government’s value added only includes salaries. However, the government also purchases a lot of goods and services from businesses, from coffee to cars, from pencils to plane tickets, to the office rentals for regulating bodies such as Ofcom. The producers of these goods and services, not the government, take credit for the value added. Since government is treated as a final consumer, the purchase of goods and services increases its spending. Clearly, government expenditure can be higher than what it charges (e.g. fees for services) because it raises taxes to cover the difference. But need the value of government be undermined because of the way prices are set? By not having a way to capture the production of value created by government – and by focusing more on its ‘spending’ role – the national accounts contribute to the myth that government is only facilitating the creation of value rather than being a lead player. As we will see in Chapter 8, this in turn affects how we view government, how it behaves and how it can get ‘captured’ easily by those who confidently see themselves as wealth creators.
SOMETHING ODD ABOUT THE NATIONAL ACCOUNTS: GDP FACIT SALTUS!
Apart from this curious view of government, the national accounts expose a number of other accounting oddities. GDP, for instance, does not clearly distinguish a cost from an investment in future capacity, such as R&D; services valuable to the economy such as ‘care’ may be exchanged without any payment, making them invisible to GDP calculators; likewise, illegal black-market activities may constitute a large part of an economy. A resource that is destroyed by pollution may not be counted as a subtraction from GDP – but when pollution is cleaned up by marketed services, GDP increases. And then there’s the biggest oddity of all: the financial sector.
Does the financial sector simply facilitate the exchange of existing value, or does it create new value? As we will see in Chapters 4 and 5, this is the billion-dollar question: if it’s answered wrongly, it may be that the growing size of the financial sector reflects not an increase of growth, but rent being captured by some actors in the economy. First, however, there are some other inconsistencies to be considered.
Investment in Future Capacity
First, let’s look at how R&D is dealt with in the national accounts. Before 2008, the SNA considered in-house R&D to be an input into production24 – in other words, a company’s spending on R&D (research equipment, laboratories, staff and the like) was treated as a cost and subtracted from the company’s final output. However, in the 2008 version of the SNA, in-house R&D was reclassified as an investment in the company’s stock of knowledge, to be valued ‘on the basis of the total production costs including the costs of fixed assets used in production’.25 It became a final productive activity rather than just an intermediate cost towards that activity.
The SNA’s decision to reclassify R&D was justified less by value theory than by ‘common-sense’ reasoning: the contribution of ‘knowledge’ to production seemed to be significant, and should therefore be recognized. R&D was made productive because it was considered important.
As a result, since 2008 GDP has been enlarged by the annual cost of R&D, including the depreciation of fixed assets used. When in 2013 the US implemented this change, the value from R&D added $400 billion – 2.5 per cent of US GDP – to national income overnight.26 Of course, those sectors with the largest R&D contributions improved their share of GDP, making them look more important than others.
The Value of Housework … and the House
Then there’s housework. Feminists in particular have long objected to the lack of recognition given to housework’s contribution to the economy. The national accounts exclude all housework, and therefore a large part of women’s work, from production. The architect of the first and second editions of the SNA (1953 and 1968), the Nobel Prize-winning British economist Sir Richard Stone (1913–91) – sometimes called ‘the father of national income accounting’ – had decided views on the matter. Writing for the UN committee that drafted the first SNA evaluation of household production, Stone commented that it ‘is unnecessary to impute an income to family services or to the services of household equipment and may even prove an embarrassment to do so, since, not only are there very little data in this field, but the principles on which such imputations should be made are obscure’.27 He simply thought it was impossible to know how to do it – and even if a solution could be devised, doing so would be socially awkward.
Now, seventy years later, since there is still no theory – beyond ignorance or shame – that explains why housewives (and house husbands) should not be included in GDP, the SNA architects have come up with a different defence. They have expressed a ‘reluctance’ to include such work because, although it is equivalent to work done by servants, ‘By convention … only the wages of the domestic staff are treated as the value of output.’28 The ‘convention’ here is ironically close to Marx’s value theory that only someone who produces a surplus for a capitalist generates surplus value. But Marx’s point was linked to his value theory and understanding of how capitalism works (or does not work), whereas in this instance the convention has been cherry-picked because it is convenient for the current system.
In explaining why housework is accounted as unproductive, national accountants are forced constantly to fall back on their ‘comprehensive’ production boundary, and are at pains to invoke ‘common sense’. Their explanations include: ‘the relative isolation and independence of these activities from markets, the extreme difficulty of making economically meaningful estimates of their values, and the adverse effects it would have on the usefulness of the accounts for policy purposes and the analysis of markets and market disequilibria’.29
According to this awkward logic, a nation would increase its GDP if we paid our neighbours to look after our children and do our laundry, and they paid us to do theirs.30 Underlying this ‘common-sense’ approach to household work is the utility theory of value: what is valuable is what is exchanged on the market. The implicit production boundary is determined by whether money changes hands for the service. Therefore, there is ‘extreme difficulty’ in giving a value to work done by women (or men) who do not receive a wage in exchange for it.
By contrast, it is remarkable how national accountants go to great lengths to include inside the production boundary the house itsel
f, the property in which the supposedly unproductive household work is done. In the national accounts, houses owned by their occupants generate services that are included in GDP. In the US, such ‘work’ contributes 6 per cent of GDP – that is, a cool $1 trillion – even though none of these dollars actually exist.
How do the statisticians come up with such an absurdity? They impute a rent to everyone who lives in their own home. A market rent is estimated for a property which the owner-occupier then pays herself as lessor for the services the house provides. Since the imputed rent is regarded as income, it is also recorded in the national accounts as production. Accountants justify this with the argument that ‘both international and inter-temporal comparisons of the production and consumption of housing services could be distorted if no imputation were made for the value of own-account housing services’.31
How might this work? Let’s contrast two countries. In one, there are only renters paying owners such as real-estate companies (in Switzerland in 2014 more people lived in rented homes than in owner-occupied homes). In the other, all houses are owned (in the US and UK a larger percentage of people own than rent). Since real estate adds value and income (rent) from the actual rent charged (as opposed to the ‘imputed rent’ calculated), the first country would have an unfairly high GDP compared to the other, at least in terms of the percentage of GDP deriving from property.
From a different perspective – one that sees no greater value in renting over owning a house, especially when there is no rent control – we could equally well ask why real-estate rents should add value in the first place. Another valid question is why a hike in rent should increase the value produced by real-estate agencies, especially if the quality of the rental service is not improving. London and New York City tenants, for example, know only too well that property management services do not improve even though rents rise – in London’s case, rapidly in recent years.32
It’s also worth noting that the national accounts treat property and real estate (both residential and commercial) as comparable to a firm. Buying a house or factory building is called an ‘investment’. It is assumed that the owner goes on ‘servicing’ the building, investing in its upkeep or improvement, so their income is ‘payment for a service’ and not just rent. Capital gains on buying and selling property are treated like those that apply to a business or a financial asset – although the extent to which a building is ‘productive’ is debatable. Capital gains from holding property arise out of increases in land value, which itself are determined by collective investment (in roads, schools, etc.) – little to do with the effort of the property owner.
As with the absurdity of neighbours paying each other to do their housework, it is as if the statisticians are saying that a nation of owner-occupiers could artificially amplify GDP by swapping homes with their neighbours and paying rent to one another. Statisticians have fiercely defended their treatment of income from property. But when real-estate prices appreciate rapidly, as in the US and the UK before 2007 and in hot-spots such as London even after the financial crisis, there are alarming implications for measuring. Rising house prices mean rising implicit rentals, and hence rising incomes when the implicit rental is included. The paradoxical result is that a house price bubble, perhaps caused by low interest rates or relaxed lending conditions, will show up as an acceleration of GDP growth. Why? Because households’ services to themselves – as their own landlords, charging themselves implicit rentals – are suddenly rising in value, and that is counted as income which adds to GDP. By the same token, if you strip out those imputed rentals, GDP can be shown to have risen more slowly in the years before the financial crash than after 2009.33
Prostitution, Pollution and Production
So national accountants’ approach to valuation affects the production boundary, sometimes in intriguing ways. In the Netherlands, where prostitution is legal and regulated, the tax authorities have asked sex workers to declare their earnings, which count towards national income. In other countries, such as the UK, earnings from prostitution are not included in national income, except perhaps in estimates of the black economy.
Equally importantly, the boundary loops around the issue of the environment. Consider a river polluted by industrial waste. When the polluter pays to clean it up, the expenditure is treated as a cost which reduces profits and GDP. But when the government pays another company to clean up the river, the expenditure adds to GDP because paying workers adds value. If the cost of cleaning up pollution is borne by someone other than the polluter it is called an externality – the cost is ‘outside’ the polluter’s profit-and-loss account – and increases GDP. Kuznets argued that such a calculation should be balanced by the ‘disservice’ that has been created by pollution, and therefore that the cost of that ‘disservice’ be taken out of the ‘net’ calculation of value added. But national accounts do not do that: instead, they state that it is not ‘appropriate’ or ‘analytically useful’ for ‘economic accounts to try to correct for presumed institutional failures of this kind by attributing costs to producers that society does not choose to recognize’.34
National accountants present this question of whether something is ‘analytically useful’ or not as a vague argument, without reference to value. To be fair, they also rightly caution that it would be extremely difficult comprehensively to cost such externalities – negative or positive ‘side effects’ of production – which are not priced. All of which just highlights the difficulties of being consistent and drawing a clear production boundary.
So while Marshall claimed that Nature does not make jumps (recall the discussion of natura non facit saltus in Chapter 2), national income, it appears, can do so! If self-employment (referred to as own-account production for small farming or sex workers, for example) grows in importance, or if a way can be found to cost externalities, national income will jump when the statisticians decide to include it.
The Black Economy Gets into the SNA
Something similar happens with the black or – to use the official euphemism – ‘informal’ economy when countries decide that it has grown so large that they must start to include estimates of it in national accounting. Consider Italy, a ‘developed’ country. The Group of 7 (G7), the international club of the biggest economies, estimates that in 2015 the informal economy made up 12.6 per cent of Italy’s GDP.35 That calculation excludes illegal activities, which Italian statisticians decided to leave out of their GDP measures. Since the Great Recession which began in 2008, many more unemployed Italians have taken up informal production. The Organization for Economic Cooperation and Development (OECD), the grouping of mainly high-income countries, estimated that in 2013 Italy’s black market (including illegal activities – around 1 per cent of GDP) was a massive 21 per cent of GDP.36 The same study found that, across other European countries, informal activities comprised between 7 per cent and 28 per cent of GDP – activities which were incorporated in the national accounts upon the recommendation of the 1993 and 2008 SNA.
All this begs the question: where does one start and stop? What is, or is not, to be included in the national accounts? The very fact that these questions are so difficult to answer illustrates the idiosyncrasies and vagaries of the accounting system. And the biggest oddity of all has turned out to be the so-called ‘banking problem’: how to estimate the productiveness of finance.
More than any other sector, finance highlights the arbitrary way in which modern national accounting decides where to draw the production boundary. When the financial sector was small (before its boom in the 1970s), there was little difficulty in excluding it; interest was as much a question of morality (positions against usury) as of economics. But as the size of the financial sector grew it became more awkward to exclude it from national output. The tension between economists’ – and indeed society’s – long-held views of banks as unproductive and the steady post-war growth of the sector gave rise to what is known as the banking problem.
Until the
1970s, one of the principal sources of banks’ profits – net interest payments, which are the difference between the interest that banks charge for loans they make and the interest they pay on deposits – was excluded from output in the national accounts. The only part of banks’ income which was included was fees for services people actually paid for, such as the cost of opening or closing a bank account or getting mortgage advice.
Yet next came an extraordinary change. From being perceived as transferring existing value and ‘rent’ in the sense of ‘unearned income’, finance was transformed into a producer of new value. This seismic shift was justified by labelling commercial bank activities as ‘financial intermediation’, and investment bank activities as ‘risk-taking’. It was a change that co-evolved with the deregulation of the sector, which also swelled its size even further. As this part of the story – how finance has been ‘accounted’ for – is too big to treat in this chapter, the next two will be devoted to it.
Profits versus Rents
As we saw in Chapter 2, the discussion about which parts of society were productive or unproductive was much less explicit before the arrival of marginal utility theory. And as we have seen in this chapter, moreover, as long as products and services fetch a price on the market, they are deemed worthy of being included in GDP; whether they contribute to value or extract it is ignored. The result is that the distinction between profits and rents is confused and value extraction (rent) can masquerade as value creation.
The Value of Everything (UK) Page 12