Car manufacturers: the case for doing nothing for the sector

Automobile producers all over the world are in dire straits. Sales and production are plummeting.  Unsold inventories are building up on dealers’ lots.  Christmas is coming too soon for many workers whose normal end-of-year break has been extended by days or even weeks.  Losses are rising fast.  Bankruptcy is looming for quite a few household names.  What, if anything, should governments do?

The US government has just announced a $17.4 bn loan to the three American automobile producers GM, Chrysler and Ford, $13.4 bn up front, with the rest coming in February 2009 – on Obama’s watch. The loan is short-term, until the end of March, 2009.  The bulk of the money, if not all of it, is likely to be drawn by the two basket cases – GM and Chrysler.

The conditions attached to this ‘bridge loan’ look tough:  “if the firms have not attained viability by March 31 2009, the loan will be called and all funds returned to the Treasury” – a step that would bring about their immediate liquidation. The White House also said that a company would only be deemed viable if it had a positive net present value, taking into account all current and future costs, and could fully repay the government loan.

Sure. The Dutch have a saying: “You can’t pluck feathers off a frog “.  When, with likelihood verging on certainty, neither the interest nor the principal on a loan will be repaid, we call that loan a grant.  The Bush administration has externalised a political internality by giving GM and Chrysler just enough cash to make it to the beginning of Obama’s term, when the fate of the US car industry becomes the new president’s problem.

The  Swedish government is about to announce financial support for Saab (owned by GM) and Volvo (owned by Ford). The German government is about to provide financial support for Opel (owned by GM).  The French government is chomping at the bit to lavish financial support on Peugeot, Citroen and Renault.

In the UK, Peter Mandelson has thus far managed to avoid giving direct financial support to the automobile industry.  He has indeed sounded remarkably sensible on the issue.  It is unlikely, however, that he will be able to hold the line.  With Land Rover and Jaguar (both owned by Tata) marching on Downing Street with their hands held out, and with the rest of the UK car industry (owned by Nissan, Ford, Volkswagen, BMW and GM) probably not far behind, the political momentum behind state support for the industry will be irresistible.

The welfare economics of public assistance for firms or industries

Let’s remember the good grounds for government intervention in an industry: (1) efficiency, that is,  the correction  of market failure and (2) equity or fairness, that is, the correction of distributional anomalies.  And’s let’s remember the bad grounds for government intervention in an industry: special interests, organised and highly vocal, trying to squeeze resource transfers or other subsidies out of the tax payer through ferocious lobbying – Mancur Olson’s Logic of Collective Action at work.

A sectional interest with much at stake, well-organised, well-financed and well-connected to those pulling the levers of political power can often enrich itself at the expense of the public interest, when the latter is represented by many citizens each of which may have a limited amount at stake.  By obtaining subsidies of various kinds, a limited number of stakeholders in the car industry (workers, management, shareholders and local communities) inflict a significant cost on the rest of society.  Because this cost is shared by many tax payers, it is quite possible that the cost of organising and mounting an effective opposition to the subsidy is not worth it for each tax payer individually.  The few exploit the many.  The same happens if protectionist barriers are erected to protect the national automobile industry.  This again benefits a limited number of stakeholders in the car industry at the expense of greater damage inflicted on the wider community – in this case would-be car purchasers.

Which reason(s) for public assistance appear to be at work, in the case of the automobile industry?

Does the automobile industry quality on efficiency or equity grounds?


The only semi-convincing fairness-based case for assisting the automobile industry is that governments all over the world are pumping trillions into their banking and financial sectors.  Those working in the banking sector are typically not worse off than those working in the car industry.  So it ain’t fair to rescue the bankers but not the car manufacturers.

Nice, but no cigar.  The reason governments should offer financial support to the banking industry and to selective other bits of the financial sector that provide key ‘infrastructure-type’ financial services (payments, clearing, settlement and associated custodial services) is not distributional.  Bankers and others working in the industry are not among the wretched of the earth who need special care packages from the tax payer.  It is despite the likely undesirable distributional consequences of the rescue of the banks that public money has to be devoted to rescuing the banks.

The reason is that the efficiency arguments are so overwhelming (the faiulure of financial intermediation has economy-wide effects way in excess of the value of the services provided by the financial sector), that we have to hold our noses and live with the negative distributional consequences.  If we could rescue the banks and the other systemically important bits of the financial system without financially benefiting those who work in the sector, who own equity in the sector, or are otherwise financially exposed to the sector (as creditors, depositors, bond holders etc.) we should do so.  But we cannot.

That does not mean that anything goes.  There always are many ways of skinning the cat.  Bailouts can and should be structured to minimize the assault on fairness that they inevitably represent.  They should also be structured to minimize the adverse incentive effects of the bailout on future behaviour (moral hazard).  Many opportunities to do so have been missed, especially in the United States of America.  Most of the rescue efforts undertaken thus far by the Treasury, the Federal Reserve and the FDIC seem designed to maximise the protection of the well-connected incumbents of the leading financial institutions, while encouraging future excessive risk taking and in violation of any reasonable notion of distributional fairness.

If the automobile industry were to be bailed out,  it is key that the extent to which this represents a reward for past inefficiency, incompetence and sloth be minimized.  The insistence by many House Republicans, that financial support should be in exchange for (1) an immediate debt-for-equity swap (for two thirds of the existing debt), (2) the immediate reduction in the number of dealerships to competitive levels, (3) a reduction in labour costs (earnings and benefits) to the level of the US subsidiaries of foreign car manufacturers) and (4) the departure (sans golden parachute) of the senior executives at all companies that make use of the public funds, seems completely reasonable.  None of these four conditions survived, regrettably, in the bail-out just offered by the Bush White House.


Automobiles are prima facie normal private goods.  There are rival in use (I buy one and use it means you cannot buy it and use it).  They are also excludable. Despite the reality of car theft and joy riding, I can most of the time exclude others from using my vehicle without my consent. They don’t create any positive externalities in use.  Indeed, as major sources of carbon dioxide emissions, they produce negative environmental and amenity externalities.

Are there any externalities involved in the production process of cars?  The automobile industry decades ago ceased to be an innovative sector, let alone one where inventions were made and innovation took place that benefited not the producers and the shareholders in the sector but the wider community.

It’s also a stretch to view cars as ‘merit goods’, that is, goods whose consumption by me is valued by others (education is often given as an example).  A better case can be made for cars being de-merit goods – ugly, space-consuming, testosterone-inflating, human-sacrifice-demanding contrivances.

The value of the production lost due to the shrinking or closure of the automobile manufacturing sector cannot be more than the value of the automobiles that were produced.  This is on the assumption that the resources that went into the production of these cars have no alternative uses, including valued leisure.  Note that, by taking the value of the cars produced (their market price at factor cost) as the measure of the value of what is produced by the industry, I include the value of the inputs produced by all the producers of components and all the suppliers of services that go into the production of automobiles, but are not employed by or owned by the producer of the finished product. Strictly speaking, if components are imported, they should not be  counted as national value added in a nationalist cost-benefit analysis.

Clearly, the notion that the resources (human, land and capital) that would become unemployed if the car industry were to shrink or collapse would remain unemployed or even become unemployable, is ludicrous. Yet this is the picture sketched by those requesting, nay demanding, financial support for the car industry.

The truth is that the car industry is cyclically more sensitive that most other industries, because it produces durable goods. The business cycle is driven mainly by fluctuations in investment demand, both inventory investment and fixed investment.  Household investment, through residential construction and home improvement and through purchases of new durable goods including cars, is an important contributor to cyclical fluctuations.  It is therefore hardly surprising that in the current downturn, much of the initial real economy pain has been in the housing sector (which had expanded excessively in any case during the housing boom that ended just over a year ago) and among producers of white goods and automobiles.

Purchases of food and many other non-durable consumption goods and of consumption services are not as easily postponed as purchases of consumer durables.  I have been testing the postponability of automobile purchases ever since I returned to the UK in 1994.  I bought a 3-year old car then and still drive that old banger today.  Bits keep falling off, but most of the time I can stick them back on again.

Manufacturers who cannot live with the normal cyclical swings in automobile purchases should not be in business.  Proposals for encouraging early scrapping of cars and their replacement by new cars are deeply irresponsible.  I’m with the Archbishop of Canterbury on this one.  Why not send squads of young thugs around car-abundant neighborhoods with sledgehammers and flamethrowers to increase the natural depreciation rate of the existing stock of cars though some targeted, government-funded vandalism?

In addition, the motor vehicle industry in the western world is set for a secular decline, as the comparative advantage in medium-tech manufacturing continues to shift towards emerging markets and developing countries.

Finally, the global large-scale motor vehicle industry is in full transition from the internal combustion engine and carbon-based fuels to battery-powered or hydrogen-powered propulsion mechanisms.   The temporary collapse in oil prices – a consequence of the global slump in activity – does not change this picture materially.  The US automobile industry, GM and Chrysler in particular, are the global laggards in this transition. They continue to produce cars that are not wanted at a cost that competitors could trump were they to produce the same unwanted vehicles.  These competitors (US subsidiaries of foreign manufacturers and foreign manufacturers competiting outside the US markets with US exports or with cars produced by foreign subsidiaries of US manufacturers) in fact often produce superior vehicles at a lower cost.

There is no economic rationale for supporting proven global losers like GM and Chrysler, firms with a record of incorrigible underperformance going back decades.  Better to put them into Chapter 11 (insolvency protection) or Chapter 7 (insolvency and liquidation) right away, with the government providing financing to facilitate an orderly restructuring or liquidation.

So can and should anything be done to support the automobile industry?  The answer has to be ‘no’ as regards industry-specific assistance.  The automobile manufacturing/motor vehicle manufacturing industry does not have systemic significance beyond what it can produce and sell profitably.

If there is a failure of aggregate demand that causes involuntary unemployment and excess capacity, the remedy is general measures to stimulate demand, through expansionary monetary policy, by boosting public spending on goods and services, by temporary tax cuts targeted at liquidity-constrained and current-disposable-income-constrained households, or by temporary investment subsidies or investment tax credits.

Governments should direct public expenditure towards those areas (infrastructure, health, education) where the social returns are highest, not towards those areas (motor vehicle production) where the special pleading is loudest.  Households will spend their tax cuts or transfer receipts where they see fit, not where Detroit or Luton see fit.

If the problem is one of credit market failure, then credit market failure in general should be addressed ‘at source’.  If suppliers cannot get their credit insured because of pathological risk aversion in the credit insurance markets, the state could consider providing temporary credit (co-)insurance for suppliers in general – not just for suppliers to the automobile industry.  If consumer credit is extraordinarily (and suboptimally) tight because even good underlying car loans, home loans and credit card receivables can no longer be securitised, then general support for gold-standard asset-backed securities (ABS) markets may be indicated.  TARP-like constructs may be able to deal with the existing stock of toxic ABS.  The Fed’s new TALF (Term Asset-Backed Securities Lending Facility) may be able to boost the flow.  Again, this should not just be directed at securities backed by automobile loans.

Instead of throwing $17.4 bn at a collection of dead ducks, the Fed or the Treasury could expand the programmes for purchasing commercial paper issued by firms in all industries, not just by the Detroit Trinity.  I can conceive of the Fed (guaranteed by the US Treasury) extending its direct purchases of private securities to include unsecured corporate loans and longer-term debt instruments issued by US corporates.  This is clearly the direction where its quantitative easing and qualitative easing are taking it.

Again, there should be no favouritism between industries.  Ice cream vendors have equally strong (weak) claims to public assistance as do the Detroit Trinity.  There is nothing special about automobile manufacturing that gives it a unique claim on the public purse.

Finally, the collapse of at least two of the three US car manufacturers (which is inevitable unless Chrysler and GM are nationalised and are allowed to live a life after death at the tax payers’ expense) would be devastating for the regional economy of Michigan and the adjacent states, and especially for the greater Detroit area.  Assuming that open-ended tax payer support is out of the question, this calamity is likely to occur soon.

The relevant policy question then becomes: should policy aim to encourage new jobs to move to where the (idle) people are or should it instead aim to encourage the idle people to move to where new jobs can be created most easily and cost-effectively?  I don’t know enough about regional comparative advantage in the US to answer that question, but it is one that has to be asked.  The regional distribution of employment and production is forever changing.  Defending the status quo without carefully investigating the cost of doing so may result in formely competitive regions that have lost their comparative advantage, entering an economically, politically and socially destructive form of limbo, in which the old industries cannot die because they are on fiscal life support, but cannot thrive because the ever-changing regional, national and global forces of agglomeration and specialisation have relegated them to backwater status.

Assisting adjustment to fundamental structural change can be a responsibility of the state either for distributional reasons or because market signals and other private incentives for inter-industry, intersectoral and inter-regional labour mobility and capital mobility are distorted or ineffective.  Assistance in retraining and retooling may be efficient as well as politically expedient. Well-choses infrastructure investments may make a difference, especially in countries like the US and the UK that have for decades suffered from a history of underinvestment in infrastructure of any and all kinds .

What not to do: the hubris of the social engineer on steroids

I want to end, though, with a great statement of exactly what not to do.  British prime minister Gordon Brown, in response to a statement by the Archbishop of Canterbury (who had likened the British government’s plans for boosting consumer demand to an “addict returning to the drug”), evoked the parable of the good Samaritan in support of the proposition that “… every time someone loses their home or a small business fails it is our duty to act and we should not walk by on the other side of when people are facing problems”.

It is no minor achievement to be a bad economist, bad theologist and bad political scientist in the space of one sentence.  Individual morality is qualitatively different from the ‘morality’ of government or the state, if the words ‘morality’, ‘government’ and ‘state’ can ever be used together at all.  The state (the government) cannot do something useful or effective every time someone loses their job or a small business fails.  This is the most awful hubris – a grotesque overestimation of what the state can do for us.

Unemployment insurance (a misnomer, really, for subsidies to unemployment) and retraining opportunities are policies that may or may not be effective in mitigating some of the effects of unemployment.  Businesses failing is a natural and inevitable occurrence in a market economy with risk and uncertainty.  It may be an individual or family tragedy, but it is not a social or economic problem, and it should not be a political issue. It is an integral part of how the machinery of a wealth-creating market economy works.

Even if the government could do something about individual cases of unemployment or small business failure, God forbid that they should try.  The responsbility of government is not to provide security at all cost – to keep us safe from all harm.  It is to defend and safeguard our freedom.  The main threat to our freedom is not Al Queda,  other mad terrorists or the acknowledged totalitarian threats of the past and the present.  It is government itself – the well-intentioned, paternalistic, we-know-what’s-good-for-you, we’ll-ram-your-happiness-(as-we-see-it)-down-your-throat apostles of benevolent despotism that are in the ascendant throughout what I used to think of as the free world.

Who wants to give me odds on the British government announcing a financial support package for at least part of the UK manufaccturing sector before the end of 2008?

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website