The collapse of Enron has shaken markets because it has reminded everyone that corporate accounts are interpretations, not facts.  Even the most conservative of accountants has  been under pressure to find statistical confirmation of the stories of heroic leadership, organisation transformation and technological revolution.  Everyone knew these things were true, even if data were sometimes slow to reveal it.

This hubris distorted perceptions not only of the performance of US companies, but also of the performance of the US economy itself. We talk about economic growth as if it were objective fact, like population growth, or temperature.   But national income accounting is every bit as much a subjective enterprise as the private sector accounting on which it ultimately depends.

When politicians and pundits talk about economic growth they are referring to movements in the level of gross domestic product at constant prices.   This concept is measured gross  – no account is taken of asset depreciation or obsolescence.  And GDP is deflated by a price index so that it represents the volume, rather than the value of output.   Such deflation was easy when output was mostly steel but is much harder in the knowledge economy.

A bar of steel is – more or less – a bar of steel:   the volume of computers is an elusive concept.   As computer prices have fallen, the result has not been that people have paid less for computers.   They have simply got more computer for their money.   But how much more?  The estimates given by different national statistics offices for the fall in computer prices over the last five years range from 30% to 75%.  Britain is relatively conservative and the United States very aggressive.

This makes a big difference.   Real expenditure on computers in 2000 in Britain was about £10 bn.   The UK’s Office for National Statistics estimates that computers which cost £10 bn in 2000 would have cost £18 bn in 1995.  If US price indices were used, the figure would be £37 bn.  The difference amounts to 2% of British GDP.   Over the five years, Britain’s reported growth rate would have been almost ½ % per year higher if UK statisticians had used US price indices.

So what is the right answer?   We are not trying to measure the benefits of computers.   These may well be much larger than £18 bn, or even £37bn, but such effects are already included in the output of the industries that use computers.

The £18 bn figure is an attempt to answer a different question.   What part of business spending on computers should be capitalised, rather than treated as a cost against current output, because it contributes to future rather than current output?  General accounting practice allows you to avoid charging such expenditures against profit with two qualifications. You must capitalise it at its actual cost, not at some hypothetical measure of what it might have cost in the past or be worth in the future.   And you must write off the capitalised expenditure over the lifetime of the asset.   The rules for measuring GDP do not impose either of these conditions.  They allow extravagant revaluation. And they don’t require depreciation of the capitalised expenditure.

Under standard accounting principles, the maximum expenditure which you could capitalise would be the whole of real spending on computers in 2000 – £10 bn or so.  And you could justify this only if you could argue that there was no need for any write down of previous expenditure on computers  as a result of scrapping or technological obsolescence.

My own estimates are that the replacement cost of the stock of computers in Britain in 2000 was probably about £20 bn.  Available computing power probably rose in 2001 by 20% or so as a result of net new investment, minus depreciation and scrapping.  Because of the falling price of computers, this larger stock of computing power was probably not worth any more at the end of the year than the smaller stock was worth at the beginning.   A kindly auditor, like the Andersen folk down in Houston,  might allow you to treat £5bn or so of computer expenditure as capitalised.  If a commercial company seriously proposed to credit £37 bn to its profit and loss – on the grounds that this is what it might have had to pay if it had not bought them so cheaply –  its directors would certainly face a congressional committee and probably end up in jail.  The bottom line of all this is that published data on GDP probably overstates output growth in the UK over the past five years, but by less than 1%.   

Economists have known for years that constant price GDP was a flawed measure of output.   But, until the information and communications technology (ICT) revolution, errors were small and offset by the advantages of data series that were comparable over time and between countries.   US statisticians were quick to see the difficulties falling computer prices might pose, and adopted a procedure called chain linking to reduce the distortions – Britain’s Office for National Statistics will follow but has not yet done so.   The problems of interpreting US accounts are both complicated and reduced by chain linking.  But the aggressive US assumptions about ICT price falls mean that the difference between GDP growth and output growth is larger in the US than in Europe.

Over the period 1996 – 2000, ICT investment contributed almost 1% per annum to reported US growth.   Simply substituting net investment at cost for gross investment at revalued prices reduces this contribution by about half.   The effect is that reported US GDP growth overstates the actual growth of  US output by around ½ % per annum over the period. This accounting difference is equivalent to the major part of the productivity miracle which still enthuses believers in the new economy.

It is not just US companies whose figures are in question.  USA Corp capitalised much of its software expenditure, revalued that expenditure at the highest price it might ever have paid, calculated its profits without any depreciation of its revalued assets, and announced stunning results to its investors on the basis of these assumptions.  

Who were its officers at the time?   Bill Clinton, the former chief executive, may be spending more time with his family.   But Alan Greenspan, who has repeatedly argued that US economic statistics should be more consistent with the optimistic reports of US business people, is still the chief financial officer.

Extended Edition:

The privatisation of air traffic control was highly controversial.  Yet when the government made a £30m loan to the business, ahead of a regulatory review to allow it to raise charges to airlines, it became clear that NATS was not a true private company.   Ordinary businesses faced with a fall in trade do not put their prices up to recover their costs:  they put prices down to stimulate demand.   That is how airlines have reacted to September 11.  The risk that demand will not grow as rapidly as expected is precisely the kind of risk that shareholders in an ordinary company expect to bear.   

But the airlines that hold a majority stake in NATS are not shareholders, but stakeholders.   They did not expect to take any significant equity risk and are neither able nor willing to do so.

NATS is simply the latest of many hybrid institutions, neither public agencies nor private businesses.   Others include old Railtrack, new Railtrack, and the special purpose vehicles that have proliferated under the public finance initiative;  restructured and thinly capitalised water companies such as Glas Cymru; hospitals grouped in quasi-autonomous NHS trusts; and the oldest and most anomalous case of all – the BBC, which is part public service, part media empire.  

Each of these institutions is different.   Each has a distinctive funding and capital structure, and a distinctive mechanism through which its management is appointed.  These differences exist not because of differences in the appropriate form of organisation for their functions, but because of accidents of history.

The problem with this proliferation is not simply administrative untidiness.   It is the resulting combination of complex legal rules and contractual arrangements combined with uncertainty about how these arrangements will work when they run into trouble.  

Who bears what risks, who is responsible for what, who is accountable for what to whom?   Mostly there are no clear answers to these questions, and even if the formal structure seems to offer clear answers they are – as in the case of NATS – overridden at the first sign of crisis.  This ambiguity in responsibility and accountability was the principal cause of the poor performance of Britain’s nationalised industries.   It led to repeated and inconsistent interventions that undermined the responsibility of managers without transferring that responsibility to anyone else.  Unwillingness to delegate responsibility combined with readiness to assign blame is  the hallmark of bad management everywhere, and is still endemic in Britain’s public services.

The best direction of reform would be a common model of a public service corporation, appropriate to the wide range of activities that need to be distanced from political control but cannot be provided by a competitive private market. 

The real issues are about management, not finance.  Begin by getting rid of  the distraction of off-balance sheet financing.  Many of these arrangements – from telecommunications privatisation to air traffic control – sprang from a desire to take the cost of public services out of public sector accounts.  This made some sense when the Conservatives in the 1980s had tied themselves to the mast of public sector borrowing targets.  Today it directly undermines the Chancellor’s reputation for provident finance.   

The similarity of title between the special purpose vehicles of Britain’s public private partnerships and Enron’s notorious special purpose entities is not coincidental.   The object of both is much the same – to exploit the accounting loophole that allows you to leave out liabilities of associated undertakings if you introduce third party equity.   

After Enron companies will now come under pressure to abandon this practice and the British government should give a lead.   The only significant economic difference between government borrowing and private sector borrowing which is secured against contracts with the public sector or repaid from levies on beneficiaries of public services, is that government borrowing is cheaper.   

The objective should not be to divide the existing activities of the public sector into public and private functions – that necessitates these complex and unworkable contracts.  It should be to fuse the different strengths of private and public sectors.   What is needed in a public service corporation is a combination of public sector ethos with private sector discipline of autonomy, audit and accountability.  

The delivery of public services is a government responsibility.  You do not change that fact by changing the formal status of the body that delivers them.  Instead you risk tension between the responsibilities of companies and directors and the reality of ultimate political control.  The wise politician will exercise the resulting authority sparingly: by making good appointments and allowing those appointed to get on with the job: by using headhunters to find management, and regulators to review investment programmes and changing structures; by insisting that everything that can be contracted out in competitive markets is; and by refusing to become involved in day to day operations.

Ultimate ownership by government means that government should be the principal source of finance – on the balance sheet. There is probably something to be said for equity type finance in which the return to investors is related to performance.  But it needs to be clear – which it is not at present – what performance criteria govern these returns.   What were Railtrack shareholders entitled to expect?

The process of audit needs to be systematic –  and involve not just financial but also performance audit.   A combination of the National Audit Office and the regulatory agencies could develop the necessary expertise.  One bizarre aspect of the current developments is that claims of commercial confidentiality are used to withhold information about privately provided public services.  Even if the plans for the London Underground are a good idea, we cannot tell.   We need more information about performance, not less.   We also need decisive procedures for bringing about management changes when audit demonstrates that these are necessary.

NATS was and is a missed opportunity to define structures relevant to the whole process of public sector reform.   The next opportunity – the restructuring of Railtrack – should not also be missed.