, Britain
In January 1990 a Victorian pumping station in Devon had a new sign screwed to the brick outside. The building was the same, the pumps were the same, and some of the mains leaving it had been laid when the town still had a corn exchange. The company that owned the lot had been on the stock market for a month.
Water is the privatisation people remember as the odd one, and the oddness is worth stating plainly. The thing being sold was not a business that had to win customers. It was an appointment to supply a region, with a duty to serve everybody in it, at prices somebody else would set.
The Act turned ten authorities into appointed undertakers
The Water Act 1989 broke up an arrangement in which a small number of regional water authorities did everything at once: supplied water, took away sewage, managed rivers, issued discharge consents and policed pollution, including their own. An authority regulating its own outfalls was a well understood problem long before 1989.
The Act separated the jobs. The water supply and sewerage functions of the ten authorities in England and Wales went into companies, and those companies were floated in December 1989. Environmental regulation of rivers and coasts went to a new National Rivers Authority. Drinking water quality went to a new inspectorate. Economic regulation, which means the prices, went to a Director General of Water Services and the office that became known as Ofwat. Scotland and Northern Ireland were not part of it.
What each company actually received was an instrument of appointment: the legal standing of water undertaker and sewerage undertaker for a defined area, with duties attached. The duty is to supply, to maintain the system, and to do both for everybody inside the boundary whether or not they are worth serving.
There is no churn in either direction. A customer cannot leave and the company cannot decline one, so every commercial instinct that involves winning business is irrelevant, and the management attention that an ordinary firm puts into selling has nowhere to go except cost.
The balance sheet was arranged before the shares were sold
The companies did not go to market as they stood, because as they stood they could not have been sold. They carried debt and an enormous capital requirement against nothing but household bills.
So debt was written off and cash was put in before flotation, and the offers in December 1989 were structured to attract small investors, payable in instalments, with incentives for holding on. What matters to somebody running the business is that these companies began private life with a capital structure designed by someone else for a purpose that was not theirs, and with a spending programme already committed on their behalf.
Prices were regulated by formula rather than negotiated. Charges could rise by the retail prices index plus a factor set by the regulator for that company, reflecting what it needed to spend. The factor is reviewed periodically, every five years, and between reviews the numbers do not move.
Most households had no meter, so bills were calculated from the rateable value of the property. Domestic rates were abolished for local taxation at the start of the 1990s, but those rateable values went on being used for water bills for years, so household revenue rested on a valuation list nobody was maintaining.
The standards were rising while the prices were being fixed
The reason the capital programme was so large is that the rules were tightening from two directions at the same time.
Drinking water standards were being brought up to European requirements, with limits on nitrates, pesticides and lead, and lead meant pipework in houses the company did not own. Bathing water standards applied to designated waters during a defined season and were judged by sampling, and a great many British beaches were not meeting them at the end of the 1980s. Coastal sewage discharged through long outfalls on the assumption that the sea would deal with it now had to be treated.
For a company in the South West that combination was harder than anywhere else, for reasons of geography rather than management. The region has a very long coastline, a small resident population to pay for it, and a summer visitor population that the network has to be built for and that never appears on the billing list. Standards are set per bathing water. Costs are shared between households. Where there are many of the first and few of the second, the bill per household is higher, and no amount of efficiency changes that arithmetic.
Underneath all of it the assets were nineteenth century. Cast iron mains, brick sewers, many of them combined so that rainfall and sewage travel in the same pipe. A network like that has an asset life measured in a century, which sets an uncomfortable benchmark: replace one per cent of it a year and you renew the whole system once every hundred years, which is roughly the right rate and more than most systems were getting. Leakage is the visible measure of that backlog, and nothing about leakage improves without digging up roads.
A monopoly earns by beating the assumption, not by selling more
An ordinary business improves its profit by selling more, charging more, or costing less. An appointed undertaker has only the third.
Volume does not help, because the customer base is a map rather than a market, and using less water is something the company is generally obliged to encourage. Price does not help, because the price is written into a determination for a fixed period and cannot be revisited because the year has gone badly. What is left is the gap between what the regulator assumed the work would cost and what it costs you.
That is a genuinely different discipline from trading. Money is made by delivering the capital programme for less than the allowance, by holding operating costs under the assumption, and by not incurring penalties. It is kept only until the next review, at which point the regulator has seen what is possible and sets the next assumption there. The reward for efficiency is temporary by design. The cost of missing a commitment is not.
Then there is the money that does not arrive. The obligation to supply does not pause because a bill is unpaid, and small household debts cost more to pursue through the courts than they recover, so the cost falls on the customers who do pay. It is the one line in the account almost entirely outside the company's control.
Where this shows up in the game
The Undertaking starts in Exeter in January 1990, a month after flotation, with a water and sewerage company, Exe Valley Water, and runs for three years. It is scored on profitable months, twenty-four out of thirty-six. The difficulty is medium, which is generous given the opening balance sheet: you begin with £40,000,000 of borrowing that somebody else agreed to, secured on assets you did not choose.
The constraint that shapes every decision is that allowed revenue is fixed for the whole run. You cannot raise the price. Capital can go into mains renewal, which moves leakage, or into treatment, which moves the pollution figure, and there is not enough for both, so underneath the pipes this is a sequencing problem. A missed commitment becomes a penalty against revenue in the following year, which is how regulation actually bites: as a smaller number at the top of next year's account.
The seed is fixed at 600008, so everybody attempting it inherits the same pipes, the same coastline and the same weather, and the scores are comparable. It runs under ironman rules: one save, and what happens happens.