No one in his right mind would lament the passing of 19th Century housing conditions. But the housing boom that has helped relieve us of their horrors now threatens to inflict its own calamities. The Green Paper 'Household growth: where shall we live?' published by the Department of the Environment in November, paints a terrifying picture of a countryside about to be eviscerated by the needs and aspirations of a growing, ageing and fragmenting population.

Between 1991 and 2016, the Green Paper maintains, 4.4 million new households are likely to form in England. Confining even half of the new housing to 'brownfield sites' (places which have been developed in the past) will be 'very much a challenge', though the department is prepared to discuss 60% as 'an aspirational target'. Room for the rest will have to be found in the countryside, both by extending existing settlements and by building new ones. The new housing will swallow up an area of rural land the size of Greater London.

The DoE's projections have been largely accepted by commentators, as a regrettable but inevitable consequence of both marital breakdown and increasing prosperity. But a close examination of both the Green Paper and the DoE's general position suggests two disturbing possibilities: first that the department has overstated the problem, and second that it is not prepared to use its powers to soften the blow.

Perhaps the most striking of the paper's assumptions is that there will be no further rise in the number of unmarried or separated people cohabiting in Britain. Between 1971 and 1991 (the last year for which figures are available), the number of cohabiting households rose precipitously, from 1.3 per cent to 6.4 per cent. The Green Paper first freezes the rate at the pre-1991 figure of five per cent, then predicts that it will stay that way until 2016. In other words, the paper suggests, nearly all the adults not legally married in 2016 will be living by themselves. It is an astonishing assertion, and my requests to the department for substantiation have so far been unmet. As people living alone account for 3.5 million, or 80 per cent, of the projected new households, this premise would, if false, undermine the paper's entire case.

The Green Paper also skates over the contentious issue of induced demand. The Department of Transport now accepts that building roads induces traffic. In a rushed and somewhat defensive argument, the DoE abruptly dismisses the possibility that the same could apply to homes: that people expand to fill the space available to them.

Moreover, its 'target' of 50 per cent of new housing on old development sites is hardly a target at all. Already, with scarcely any government help, 49 per cent of new housing is built on brownfield sites. There is some dispute about how much derelict land exists in Britain, but a survey by British Gas Properties estimates at least 200,000 hectares, in theory more than enough to accomodate all the projected new homes, though not necessarily all in the right places. If this is correct, the department's 60% 'aspirational target' looks very weak indeed. The Green Paper underplays the potential for office conversions, which, in the Borough of Westminster, for example, now account for 20 per cent of all new housing.

In truth, the major disincentive to building affordable housing in cities is not, as the DoE suggests, the cost of decontamination, but a factor the report does not mention at all: the speculative value of the land itself. The market in urban land has been superheated by the laxest planning regime Britain has seen for 30 years, as well as an extraordinary array of special privileges for developers. Unlike the rest of us, they have a right of appeal against local authority decisions, which they use to threaten councils with enormous legal costs if their applications are turned down. If blackmail doesn't work, they can try bribery: the principle of 'offsite planning gain' is now so loosely interpreted that developers are simply buying planning permission.

These and other democratic deficits mean that the construction industry can impose the developments it wants, rather than those that local people desperately need. This is why derelict land is used to build superstores and service stations, office blocks and executive estates, rather than affordable homes. Brownfield land in London is now selling for as much as £3 million an acre: there is, in other words, no chance whatsoever of good cheap housing being built there.

The solutions to the purported crisis are so obvious that one cannot help but be suspicious of the Green Paper's failure to propose them. There is no mention of a redemption of developers' inordinate priveleges, or of planning guidance that would bring down speculative prices by insisting that derelict land is to be used for affordable housing, not superstores. It fails to suggest time limits, followed by compulsory purchase, to determine how long developers can sit on derelict land. Never does it pose the prickly but urgent question of making the ownership of second homes more onerous.

Together, such measures would draw the environmental sting from new house-building, whether or not the department's apocalyptic projections are accurate. Even so, some new rural homes will have to be built, not least for local people. But if city land is used effectively, it is conceivable that we could meet the need for rural housing with no more damaging developments than small increments to existing villages.

But the truth is that the last thing we can expect from the DoE at the moment is a positive and proactive approach. Building is the only industrial sector in Britain with a government agency, the Construction Directorate, specifically dedicated to promoting its interests. And it belongs, astonishingly, to the Department of the Environment. Senior figures in the construction industry are regularly seconded into it, and swing enormous weight within the department. There is no question about what they want. The House Builders' Federation bitterly contests any prospect of having to concentrate on brownfield land: it is easier and far more profitable to build in the countryside.

The Green Paper's flawed terms of reference begin to look less mysterious. The DoE is asking the wrong questions in the earnest expectation of getting the wrong answers. If the countryside is destroyed, it will be destroyed not for our sake but for that of the developers. There is no prospect of saving our most precious places until the government is brave enough to refute their insatiable demands.