Housebuilding Freakenomics

The government aims to increase housebuilding in the hope that this will improve affordability. In the long term, they might be right, but I doubt it. In the short term, the terms on which mortgage finance is obtainable might be more important. However, in any case, the supply of suitable land is a concern.

My intention here is to illustrate the extent to which the increase in land values consequent on high house prices mainly benefits landowners, when it could be used to improve the supply of affordable housing. 

The key distinction here is the difference between the value of the land in its current use and its potential value if redeveloped for housing. The increase is often enormous, and obtaining planning permission is the key to unlocking it. 

If a site has already been zoned for housing in the Local Plan, development is effectively agreed in principle, so what remains is haggling over detail, and the sale agreement can be relatively straightforward. If it has not, there will often be an extended negotiation with the local planning department and between the seller and buyer, and the outcome will be uncertain.

Candidate sites are usually either ‘greenfield’, i.e. currently farmland or ‘brownfield’, which is land that has been built on before but is no longer in use. Developers normally prefer the former, but public policy wants to prioritise reuse of the latter. 

Property developers need land. If they don’t have any, they will be out of either a business or a job, so there is keen competition to buy promising sites. A bid might succeed because it is based on a better scheme, but more often it simply reflects optimistic assumptions about the likely costs and sales receipts. Since the highest bid, by definition, exceeds the consensus, that optimism will often prove to be unjustified. This is intrinsic to any auction and known as the ‘buyer's curse’. (Or, colloquially, the Curse of Ebay.)

At this point, the developer will often seek to improve the viability of the scheme by going to the Town Planning Dept. at the Local Authority to argue that the planning requirements are too onerous, perhaps in respect of the proportion of affordable housing, infrastructure and open space. The Town Planners also have targets to meet, so they usually compromise. (Some developers keep two calculations, a credible but ‘worst case’ scenario to show to the planners, and the ‘base case’, for internal use, and which includes potential profit-maximising tweaks to the estimated costs and sales values). 

While these negotiations can restore the developer’s anticipated profit, they have the unfortunate side effect of advertising to landowners that the planners’ policies are not set in stone. This assumption then feeds into bids made for sites in the future, so the benefit of the planners’ flexibility is reflected in the price sought for land in the future. I fear this is happening in London now. The Mayoral decision to speed the pace of development by reducing the requirement for affordable housing might well help existing schemes, but the future beneficiaries will be the landowners. 

London's Mayor 

Overall, the housebuilders themselves do not make an excessive profit. If they did, their shares would be expensive, while in fact they are quite cheap and fluctuate a lot. 

In contrast, the sellers of farmland for housing have done startlingly well over the past fifty years or so. The rise in the price of greenfield land has far, far exceeded the rise in the price of the houses built upon it. And they have rarely done anything to earn their good fortune, the location decisions of employers make a difference, and roads and services that make a site valuable have invariably been paid for, directly or indirectly, by the public. This isn’t a radical observation; it was memorably pointed out by Winston Churchill a century ago, who observed of the landowner that he "renders no service to the community, he contributes nothing to the general welfare, he contributes nothing even to the process from which his own enrichment is derived". 

The scale of the benefit will vary a lot, not least because every site is different, but it is worth illustrating.

In the South East of England, where the demand for housing is most intense, farmland might fetch, say, £25000 per hectare. (NB. 1 hectare = roughly 2.5 acres)

Here is a link to data on farmland prices: Farmland Values

And here is a rather dated guide to residential development land prices, which, as you can see, are markedly higher: Housing Land Values

Note that the value of brownfield land is less predictable. There is often a need for demolition, decontamination and other groundworks, but they also tend to have better access to road and utility networks.

Why have the prices of housing land and houses not risen in tandem? To grossly simplify, over the very long term, given the supply constraints, you might expect the amount that people can pay for a house to rise with their increasing ability to pay for it, i.e. the rate of income inflation. In fact, the shortage of housing, in particular in some of the most sought-after areas, normally those where there are lots of jobs, means that prices can run ahead of this.

Meanwhile, construction costs might only rise at the rate of price inflation, which is usually lower. (If it wasn’t, in technical terms, Real GDP growth would be zero and we would be stuck in a 20th century economy). Of course, it isn’t as straightforward as that. An increase in wages also affects the builders, but on the other hand, they have had years to learn how to do things more efficiently, and with the blessing of the planners, they have increased the number of houses on each site. In developments over the past few decades, you rarely see impressive suburban-style front gardens or wide roadside verges.

One cost that has increased is meeting planning requirements, and in that a (usually contested) proportion of the houses are ‘affordable’ on either a rental or assisted sale basis. The developer makes little or no profit on those.

What a developer can pay for land is effectively arrived at by subtracting the estimated cost of a scheme and the basic profit required to justify the investment from the amount you expect to be able to sell it for. The ‘land price’ element is usually the smallest of the three figures and hypersensitive to changes in the other two. In particular, a small change in the assumed sales receipts can lead to a large change in the amount of money available to pay for the site.

For instance, imagine you see a site that is ripe for development is planned where the houses might in total sell for £10m. Your initial calculation suggests that the amount that it would cost to build it, plus the profit needed to make it worth your while, is £8m. The amount you can afford to pay for the site is thus £10m minus £8m, i.e. £2m. 

Maybe your instinct is that it will not be enough for your bid to be successful, so you reconsider those numbers to see if they can be improved. If you can find a way of reducing the costs by 5%, the equation becomes £10m sales minus £7.6m costs, i.e. £2.4m. That 5% cost saving allows you to increase your bid by 20%.

Alternatively, you could keep your original cost estimate and look again at your estimate of sales values, perhaps taking into account more recent sales from other local schemes. If these suggest that your houses might sell for 5% more, i.e. £10.5m, the equation becomes £10.5 minus £8m, i.e. £2.5m, and you can increase your bid by 25%. 

This shows just how sensitive the value of land for development is, to the assumptions made about the likely costs and sales values. Small changes in either make a big difference to the price you can afford to pay for the land. It also means that in places where house prices are highest, land values are also highest and, once again, the differences can be spectacular. 

In the simple calculation above, if the planned development was located where houses were worth twice as much (i.e. £20m), the land would be worth six times as much. (£20m minus £8m = £12m). As Private Eye magazine would say, ‘kerchingg’!

A more important consequence is that, over a long period, the difference between the rate of increase in house prices and the rate of building cost inflation, which is usually lower, leads to a much greater increase in the value of the land.  This is the arithmetic miracle of compounding, which, in the Southeast in particular, is more of a curse and makes the enormous rise in the price of land understandable. It benefits the landowners enormously, even though, to repeat the point made by Churchill, they have done nothing to earn it.

Of course, the Government is aware of this, and for a long time has sought to harvest some of the gains for public benefit. Hence, the growing scope of the required planning contributions. These are embedded in local plans, but for a long time were usually negotiable.

This raised a new problem. The basis of negotiation was usually an assessment of the viability of the proposed scheme. The Government have placed faith in the ability of valuers to correctly estimate the value of development land, simply because they need to do that to justify the policy approaches they adopted. That faith was ill-founded; in fact, there is no reliable way of doing it. The professionals are aware of this, but the general view is that it is better to have a poor method than no method. 

As explained, the value of the land is acutely sensitive to the assumptions made about how much a scheme will cost and how much it can be sold for. Up to the point at which a scheme is agreed in detail, those assumptions are essentially subjective, and there is plenty of scope for an argument in which the developer will have more expertise and information and thus the upper hand. 

Again, this problem was recognised, and the policy response was to standardise the planning charges through mechanisms such as the Community Infrastructure Levy. Or CIL. But the problem with ‘one-size fits all’ solutions is that these tend to be based on the lowest common denominator. Once again, in the short term, that benefits the developers who own their sites, but more critically, in the long term, all owners of future development land who, to repeat the point ad nauseam, have done little or nothing to earn this good fortune.

What more could be done? Taking all of the land into public ownership has always seemed draconian and expensive, so occasional efforts have focused on the more targeted solution of effectively nationalising development land or taxing away excess profits.

Recent thinking has focused just on land earmarked for development and the (unearned) increase in value that follows. One approach is that, if it could be acquired at a price much closer to its value in its current use, for instance, as agricultural or wasteland, the surplus created by the development could be diverted for public benefit, for instance, by creating more homes to rent.

This is not an uncommon arrangement in Northern Europe, but it inspires howls of outrage in England, which has a long history of protecting and nursing landowners' interests and the rules governing compensation when an effort is made to buy it for the public good are, to put it mildly, unhelpful. Their origin is, in part, bizarre. If you are interested, see my post here: Pointe Gourde

It is no accident that the Royal Institution of Chartered Surveyors (RICS) is the world’s largest body of property professionals. You might expect to hear unfounded claims that compulsory purchase runs contrary to the European Convention on Human Rights. It doesn’t.

It would require changes in the powers of public bodies to acquire land, providing them with the resources to do so, and changing the rules governing the payment of compensation for the land compulsorily acquired. A typical scenario would see a Local Authority or other agency setting out the requirements for a new scheme and then selling the sites to a housebuilder on a competitive basis, using the sale contract to lock in the public benefits.

This has the great advantage of being self-funding. Everything paid for the land should be recoverable, and more, when reselling it. But of course, there is a lot of public sector effort involved and, critically for a government worrying more about spending today than benefits tomorrow, this expenditure might precede the income by quite some time. For this reason, schemes have been slow to take off.

So, where have we got to with all this? During the New Labour era and the coalition government that followed, the emphasis was on the tariff schemes referred to earlier, oblivious to their technical shortcomings. After that, Theresa May’s government considered ideas that were more limited but otherwise similar to those outlined here. They did not survive her tenure.

May. "Forward Together" 2017 

In their manifesto for the last election, Labour said they would reform compulsory purchase compensation rules while aiming to compensate landowners fairly, rather than on ‘inflated prices based on the prospect of planning permission’. Since then, they have launched a consultation, which, in the tradition of the Civil Service, will take a long time, cost a lot, and resolve little.

Labour 2024

Meanwhile, arms will be twisted by professional and industry bodies, such as the Landowners Alliance and the RICS, who routinely cloak their ideological preferences with claims to neutrality. A year on, it hasn’t got much further, and the chances of seeing the fruits of progress before the next election will quickly shrink.

What can we expect? Frankly, I have no idea what the future holds for farming, but I would happily bet that owners of land which could be used for housing will be sitting pretty for years to come, and that this Government, like its predecessors, will fail to meet its housing targets.