Banking on the taxpayer

Third in the list of this week's bad policy ideas* is the revival of talk of a state-owned or -backed infrastructure bank. The FT wrote approvingly of how both major UK parties are considering this option:

By reducing the risk to investors, it could bring down the cost of capital for the industry and hence the ultimate cost to consumers... 

The bank could fund projects such as nuclear plants and wind farms, spreading the risk over a range of investments and issuing bonds that could carry tax advantages and possibly a state guarantee...

In return for the cheaper funding, the industry would have to accept tighter regulation, moving away from the UK's free market for energy towards a framework of returns agreed between companies and the regulator.

The Conservatives' New Economic Model, launched on Tuesday, promises:

We will create Britain’s first Green Investment Bank, which will draw together money currently divided across existing government initiatives, leverage private sector capital to finance new green technology start-ups and back the bright ideas of the future. Lord Stern has agreed to advise us in the creation of this Bank. 

It turned out that Lord Stern had agreed no such thing, thank goodness (haven't the Tories noticed that fewer and fewer people think the Stern Review's approach to carbon valuation holds water?). But that's just a sideshow. The point is that, with or without Stern, the Tories are quite committed to the concept.

Meanwhile, Labour (according to the FT) is:

considering such a bank as a way to raise funds from long-term investors such as pension funds

It looks like most of our intellectual class thinks we need goverrnment-managed lending to deal with the difficulties of long-term investment in infrastructure. They are wrong. We need to deal with the reasons that businesses judge there to be insufficient reward to justify the risk of investment in infrastructure, not preserve those inefficiencies by hiving off a chunk of the risk to taxpayers, pension-fund beneficiaries and the like. The following is an attempt to explain how and why.

Excuses for state-backed financing

Long payback periods

Long-term investment is required for the production of most goods and provision of most services, not just for energy, water and roads. Some investments may project a lifespan of 20 years, others 40 years or more, but (as will be explained below) rational investors will expect the investment to make a decent return over the first 20 years, regardless of how long the investment might theoretically last. The expected longevity of some energy investments is therefore not a substantial reason why financing should be materially more difficult and require state intervention.

Thanks to the reality of a number of factors (time-preference, risk, uncertainty, growth, depreciation, etc.) that are reflected in the accounting practice of discounting cash flows in financial models, it is the early years that count most for projects of any length. Year 1's earnings count for more than Year 2, and so on.

State involvement in financing may reduce one aspect of risk - sovereign risk (i.e. the risk of being harmed by changes to government policy) - although it is not safe to assume that this risk can be eliminated in this way (governments do not have complete freedom of action, and may be faced with dilemmas where harm to one or another interest is inevitable). The involvement of the government in the provision of finance makes no difference to the other aspects that are taken into account in the discount rate. Governments may set arbitrarily low discount rates, but that is because they are ignoring reality (acting as bad investors on behalf of the taxpayers, and probably crowding-out private investment), not because they have somehow reduced the significance of the factors that make it advisable to discount future cash flows. Even with government-backed finance, therefore, one ought to assume a realistic discount rate.

Let's say we take 6% as a discount rate that assumes as low risk and cost of capital as even an optimistic politician might reasonably assume. Even at this low rate, the effect on future values is significant. For a 50-year project earning identical (inflation-adjusted) earnings every year, once discounted at 6% a year, nearly three-quarters of the value is earned in the first 20 years, and the final 30 years contribute only one-quarter. If the project is to make even a modest return on investment, it will have done so after 20 years, and adding more decades to the financial model's time horizon will add only a little to the return (around 1.5% if extended to 30 years, another 0.5% to 40 years, and another 0.2% to 50 years).

In other words, however funded and however long the lifespan, projects need to earn enough money in the first 20 years to have made a decent return on investment. Projects with lifespans longer than 20 years need to be priced to recover their capital almost as quickly as projects with shorter lifespans, but (having recovered the capital) should be able to supply their product in their later years at a reduced price based on the marginal cost of operation.

So if infrastructure projects need to be modeled on similar returns over similar timescales to investments in other sectors, what makes them so special that they need state-backed investment, and what is it that the state-backed investment is providing that wouldn't be provided by the market?


Interventionists often like to label as "essential" whatever it is that they think the market is not delivering adequately and whose delivery therefore requires government assistance. The semantic implications of the word "essential" seem in their mind to justify government intervention. This is never a strong argument, and no more so in this case. Clothes are essential, yet we do not need government intervention to ensure their production. In the energy sector, oil for transport is fairly essential to our current way-of-life, and yet it is provided mostly by private companies and investors in the West. Much of our food is provided by the private sector, and the action of govermnents in the food sector acts mostly to encourage the wrong amount of production. We cannot easily get by without transport vehicles, which the private sector has proved perfectly capable of supplying (in fact, over-supplying) without recourse to government finance. Construction materials are neither provided nor financed by the state. Essentiality is a poor justification for intervention.

Scale of investment

The scale of the investment required is no good excuse either. Private oil companies manage to raise private finance for large R&D and extraction programs, as do chip and car manufacturers. There are huge amounts of money apparently sloshing around out there, looking for homes, judging by the bond markets (OK, that owes significantly to QE, but all the same, once the money is out there, it could as well go to financing private infrastructure investment as to propping up profligate governments - the people who think government needs to stick its fingers in yet another pie because of supposed difficulties with finance are the same ones who believe QE ensures availability of enough money in the economy). What influences appetite is not size or sector or lifespan, but risk and reward.

The real reason why the reward often does not justify the risk of investment in infrastructure

The truth is that government failures (bad interventions and failure to protect competition) have made the balance of risk and reward in energy infrastructure unattractive. The interventionists assume their measures must be justified, so if the market response to these interventions is to run away from investment, we must need yet more intervention to ensure the necessary investment happens.


For example, the EU-ETS is structured in such a way that marginal differences in demand and supply will have large effects on the price. So long as the market believes that emitters covered by the scheme will come in below target, the price of EUAs is too low to justify investment in almost any non-fossil-fired technology. Should emitters at any point find that their combined emissions are above target, the price of EUAs will leap to a level where investments in fossil-fired technology will be powerfully devalued (and maybe even be uneconomic to run). Given the number of parties to the EU-ETS, all of whom have incentives to obscure their emission prospects, not to mention the roles of politicians and political gaming in setting relatively short-term targets, and of consumers and general economic circumstances in determining the level of savings that have to be made, no wonder it is almost impossible for investors in energy infrastructure to know which side of this unstable balance they should be betting on. In these circumstances, the safest policy is to stay out. After all, this creates a potential shortfall that will drive up prices and profits for incumbents, so they have a lot more to gain from a conservative investment position than from taking a punt on the future.

Sovereign risk (micro-management, Eurocracy, regulation, grants, obligations, etc.)

We have similar levels of sovereign risk in most of the many interventions by European governments in the energy sector. The micro-managing philosophy that dominates all the main political parties in Britain (and probably in the rest of Europe) results in mechanisms constantly being tweaked to try to make the outcomes more like politicians think they ought to be, but in ways that could not have been predicted by people who invested on the strength of earlier versions of the schemes. European legislation is often forced through (without recourse), bringing dramatic changes to the economics of one or another technology. Excessive regulatory powers hold the sword of Damacles perpetually over all businesses subject to those regulations and the arbitrary powers and inconsistent judgments of their enforcers. Grant schemes carry the threat that, at any moment, a competitor gaining a substantial award may be able to develop a competing installation with whose economics you cannot compete, because of the artificially-reduced capital costs. Obligations work the same way as the EU-ETS, threatening large price-swings from minor differences in the behaviour of vast numbers of unpredictable participants.

VILE-company oligopoly

Then there is the dominance of a handful of VILE companies. One of the most important drivers of investment is the threat of new entrants coming in to fill the void if incumbents do not invest to maintain their position. The dominance of a handful of vertically-integrated players acts as a massive barrier to entry (on top of the risks that VILE and non-VILE companies alike must face). A new entrant in the generation business knows that most of his potential customers are also his competitors in the generation business, and can therefore reasonably expect not to get as good a deal from them as they give to their in-house generation (as 100% of the profits from in-house generation will be captured at whatever internal price is set). Likewise for new entrants in the supply business.

The VILE companies argue that the vertical-integration reduces their risks, but when it comes to new investment, it significantly increases the risk. In a disintegrated market with no vertically-integrated players, new generators must concern themselves only with the overall levels of demand and supply. This is the sort of judgments that businesses in all sectors have to make. It does not matter if one supplier gains market share from another. So long as the overall size of the market does not change, generators will simply supply more to the grown supplier and less to the shrunken supplier. A VILE investor in a market dominated by VILE companies faces a more difficult calculation. Most of the output from a new generating station will be intended for the customers of his integrated supply business. He does not want to have to sell his output in the market, because he will have to sell to his competitors. He therefore has to judge, not only the overall levels of demand over the lifetime of the new generating station, but how his share of the supply market may change. That depends very much on the choices of the small number of VILE competitors, which may not be entirely predictable. It is safest only to invest in generation capacity sufficient to supply a proportion of his current and projected level of demand. But that relies on independents investing to produce the balance of power required to meet demand, and we have created a market in which independents are severely disadvantaged, and therefore disinclined to invest. Hence, a capacity gap is a likely outcome of a VILE-dominated energy market.

Vertical-integration also dramatically reduces liquidity in the traded market. Most generation is contracted in-house, and most of the rest is contracted in bilateral trades between the VILE companies. These are opaque transactions that do not facilitate price-visibility in transparent exchanges, and the development of derivatives that can be used to give confidence to financial modeling exercises (such as forward prices) and reduce investment-risk (such as futures and options markets). There is a reason why futures markets in electricity are so much shorter (barely more than 3 years out) and less liquid than for those in gas and oil. The only ways to create the liquid traded markets that are essential to reducing investor-risk are either to force all electricity to be traded through a Pool (with the associated disadvantages that we had tried to get away from), or to disintegrate the VILE companies. Given the need also to reduce the VILE companies' influence over government to reduce the rent-seeking pressures on government to perpetually micro-manage their schemes to deliver outcomes that the VILE companies advise are optimal (and which just happen to favour their business models), by far the better option is disintegration.


All of the above, and the many other disincentives to investment created by government, could be overcome if the price were right. Risk does not mean that people will not invest, but that they will need the promise of faster and higher rewards to justify their investment. The real problem with energy investment, and the reason why people look for the deus ex machina solution of a state-owned infrastructure bank, is that governments pile on the risk, but expect energy to be priced as though it were the lowest-risk investment conceivable. In fact, they really want investors to assume an inconceivable level of risk - so low that earnings in later years (even 30 or 40 years out) are treated as valuable as earnings in Year 1. They do so, because the only way that they can get the unfeasibly-low electricity prices they believe should be delivered despite the need for massive capital investment, is if the cost of that capital can be spread evenly over 40 or 50 years (or even 60, in the Neverland of some nuclear advocates).

But as we explained earlier in this article, that is simply not a rational approach to investment risk. Energy investments, like any other, must be enabled to price their product at a level that ensures an adequate return within 20 years. Whether the funding comes from private sources or from a state-backed institution does not make any difference to this reality, but may make a big difference to the possibility of foolishly ignoring this reality.

The effect of a state-backed infrastructure bank

Under-pricing of risk and mis-direction of investment

The effect of funding large energy projects from a state-backed infrastructure bank would not be to reduce the risk, but (if the bank provides a high proportion of the total funds required at a sub-market rate) to move the risk from the developer to the fund-providers. If the money is provided directly by the government, it moves the risks onto the taxpayer. If the money is drawn from low-risk institutional investors such as pension funds and insurance companies, who would not invest in these types of investments for these levels of reward without the involvement of the state as guarantor, then the risk falls again on the taxpayer as ultimate underwriter of government guarantees, and also on the institutional investors and their beneficiaries (e.g. pensioners), who could find themselves holding over-valued, under-performing assets if the state guarantee turns out to be less solid than they had believed.

It sends shivers up my spine to think that people's pensions might be dependent on investments in offshore wind farms or carbon-capture and storage (CCS) projects. What a magnificent way to destroy yet another chunk of what little capital we have left.

Favouring of incumbents and big players

Our future energy systems do not have to be based on monolithic units. The optimal outcome will probably be some balance between diverse, local, small-scale units and large units enjoying economies of scale. But you can bet that a state-backed bank will strongly incline to big projects, partly because it is more convenient for bureaucrats to deal with fewer, larger units, and partly because big toys appeal to politicians who want symbolic projects that signify politically. Even if they want to back smaller solutions, a state-backed bank will not have the tools to make good judgments about which projects to back. Technologies will be favoured that are politically-preferred due to successful lobbying by one or another group, and companies will be preferred who are suitably complimentary about the government's interventions in energy and finance. The EIB demonstrates the point.

Embedding of bad policy

Once large amounts of money have been invested in certain projects and technologies, backed by government guarantees, it will be a disaster if those projects and technologies were to turn out to be uneconomic. But the economics of projects and technologies can always be changed by fresh government interventions. Faced with (a) having to admit the error of massive misdirection of funds and coff-up on guarantees on bad risks, or (b) providing additional subsidy so that the guarantees are not needed, governments will almost certainly prefer the latter option, particularly since the approach to assessing the validity of interventions has become so corrupted that almost anything can be justified. We could well end up with the tail wagging the dog - additional subsidies being poured into technologies that we have discovered are ineffective and expensive solutions, simply to avoid having to face up to this reality.

Market signals distorted

Prices serve a number of functions. They influence demand as well as supply. If a state-backed bank succeeds in its raison d'être to suppress the cost of capital invested in energy infrastructure, and therefore the price of energy to consumers, we will have more energy production and consumption than would occur without intervention. In other words, more will be invested in energy-production and less will be invested in energy-efficiency than would be the case in an efficient, undistorted market. We will be burning money supplying energy that we should instead have saved.

Silver lining

A state-owned infrastructure bank would be a disaster, and should be resisted by all means possible. But should it be foisted on us, there is a silver lining. The FT reports that:

Criticism is likely to be strongest from private sector bankers, who would be threatened by a state-backed competitor in a core business.

State intervention in, and then ownership of ever more parts of the economy is the logical culmination of the corporatist and managerialist view of economics and politics espoused by most of our intellectuals (in the Hayekian sense), including politicians and civil servants. If we diminish the roles of competition, individualism, entrepreneurialism, etc. and focus on the economies of scale that the corporates supposedly bring, we have removed most of the arguments for private ownership. The state can drive economies of scale more aggressively than any private organisation, if size is all it takes. That applies in the financial sector as much as in any other. The banks and other financial institutions will crash and burn with the rest of us whose role has been undermined by the corporatist trend. And who were the lead cheerleaders for this corporatism? That's right. The financial corporations will have destroyed themselves along with the rest of us. Again.


* The first two were the granting of civil powers to the EA and the proposals for the micro-gen FIT.



What is really gobsmacking is that the government is borrowing money (selling bonds) at 3+% and lending it out at 0.5%

And they say this will fuel a recovery