Do not forget supply as a driver of real estate returns. If you like the look of a market make sure you know its planning regime and the associated risks, says Paul Cheshire

As an economist I hammered on for 25 years about how real estate is an investment asset and needs to be treated in a strictly comparable way, and analysed with the same rigour as expected of other investment assets.

Now I teach real estate economics and finance to trained economists and find myself increasingly emphasising that actually real estate is a bit different from other investment assets and it does not make sense just mechanically to translate techniques applied to financial assets to real estate without thinking carefully about the implications of their differences.

There are three essential features of real estate we have habitually failed to reflect on sufficiently. First, unlike equities or bonds, real estate ultimately commands a return because people or firms think occupying it will help them and their businesses make money (or in the case of houses, give them welfare - but I will ignore housing in what follows although there are strong parallels).

Second, its supply is influenced by very different factors to other assets; apart from its durability, which means most of the supply of real estate assets is already built, there are powerful regulatory constraints on supply which do not affect any other type of asset.

Third, as any real estate professional will tell you (without necessarily having reflected on all its implications) real estate assets are fixed in space in a way no other assets are. When we analyse what drives returns from real estate, therefore, we have to look at what drives demand for commercial space in the relevant local market; and what controls local supply. The relevant ‘local' will of course vary according to type of use.

Analysts of retail space have sophisticated processes for identifying such catchment areas but offices - even industrial space - have relevant local market areas, and income and population growth within them will be the drivers of demand.

But suppose we play a thought experiment and imagine a world in which the supply of space is perfectly elastic. This would mean that, with costs constant, a small increase in price would generate a huge (strictly, an infinite) increase in new space as profitability sucked in more resources to construction.

In such a world, however big the increase in demand, prices - at least at the margin of development and allowing for lags - would not rise. The rents of more favourably located space would rise given the natural limits on the supply of favourable locations, but supply would expand so the price of marginal space was always the same and reflected the costs of building it.

The point of this ‘thought experiment' is to demonstrate how far we actually are from such a situation of perfectly elastic supply. One of the unconsidered consequences of real estate being so place-specific is that people who work with it and think about it are very knowledgeable about the conditions they face and the institutional, legal and planning constraints they live with in their markets: but almost clueless about the very different constraints in other markets.

The internationalisation of real estate investment of the past 15 years or so may be starting to erode this but it is still the norm. Real estate professionals tend to assume that the rules with which they have to live in their own local markets are somehow the universal state of the world. But emphatically they are not.Everything that regulates real estate varies dramatically between countries and sometimes more locally even than that.

The two most obvious factors regulating the supply of new real estate are the structure of tax incentives and land use planning or zoning systems. Just take two systems to illustrate this. In the US taxes on commercial property are imposed by, and revenues accrue to, the local community.

This will also typically be the zoning authority. Local property taxes on commercial property yield more revenue than the costs of providing public services so there is a handsome incentive for local communities to allow - even set out to attract - new commercial development. Zoning systems are local and, at least for commercial property, flexible. In most large cities there is a ready supply of new development sites or greenfield land.

In the UK the land use planning system and the property tax systems are essentially national. New commercial development is subject to tight restrictions which allocate the areas within cities within which commercial development is allowed, and set tight restrictions on heights and total space. In addition, in most cities, there will be significant conservation requirements limiting what changes are possible to existing structures and protecting sight lines or views.

There are eight protected sight lines of St Paul's Cathedral extending up to 10 miles; in most of London's West End it is almost impossible to change the external appearance of any building. Taxes on commercial property account for nearly 5% of government revenues and are a national tax. But still local communities have to provide the public services commercial buildings require. In effect, in the UK, we have created a world in which local communities are fined for allowing commercial development. Not surprisingly planning authorities in the UK are highly restrictive in their approach.

The upshot of this is that the most important influence on the costs of commercial space (and so ultimately on the returns) is the regulatory regime local developers face: regulatory regime here includes the whole bundle of legal requirements, fiscal arrangements and, above all, planning or zoning regimes setting the bounds on the development process.

After all our thought experiment has already shown that if the development process is competitive (as it seems to be), then in an unregulated world supply would be perfectly elastic and space costs would just reflect building costs.Yet comparing Manhattan and Birmingham in 2004 reveals intriguing discrepancies. The cost of constructing a m2 of office space was about twice as much in Manhattan as it was in Birmingham - no surprises there. Where there was a surprise was in the price of that space.

In Birmingham that was about 44% more per m2 than in Manhattan; a major surprise until you think about the mounting evidence on how restrictive the British land use planning system is (for example, Cheshire and Sheppard, 2002; Barker 2003, 2004, 2006a & b).

This suggests a way in which we can systematically estimate the costs being imposed on office space by regulatory restrictions. Given that development is competitive, then profit seeking would mean providing space until the costs of building it (including some normal profit) were equal to the price. If in any local market the price was higher than the costs of building a bit more space of a given type then some developer would step in and make some money by building it.

Unless, that is, they were constrained by regulation. So we can think of the difference between the costs of constructing an additional m2 of office space and its price as a measure of the locally ruling costs of regulation. If we express that difference relative to the construction costs we can think of this as a ‘regulatory tax' imposed on development. There are complications that could be added (for detail, see Cheshire and Hilber, 2008).

Courtesy of Davis Langdon's remarkable historical data on building costs and CBRE's data on prime rents, helped with data for international locations from Gardiner and Theobald and Jones Lang LaSalle, and with data on voids and rent free periods from various other sources, we were able to construct estimates for 22 major office locations in Europe and also, in a restricted way, for Manhattan.

Regarding British locations we had data going back in some cases to the 1960s, giving a total of 480 observations. This allowed us to do some interesting analysis to identify the extent of the disincentive for development moving to the uniform business rate imposed and how fear for local job prospects was the major determinant of the variation in the severity of restrictions. The table shows some summary information for 1999 to 2005 for all the markets we could get estimates.

These may look like small numbers but they are proportions of construction costs. So a value of 8.09 for London's West End implies a ‘tax' on construction costs of 809%. Just in case anyone should misinterpret, I am not arguing for a world in which there is no regulation of what can be built where. Markets need transparent regulation to operate effectively (as we have recently been witnessing so dramatically) and land markets are particularly prone to problems of so-called market failure.

But there is a sharp distinction to draw between regulation and the restriction of supply. It is that last effect we are trying to measure. We are not measuring the value of any benefits restricting supply may yield: just a reasonable indication of the gross costs.
The costs we impose via regulation of development are important in economic terms but also have implications for investors. Increasing the cost of commercial space increases the costs of providing the services that use it.

Unless space can be costlessly substituted out of the production process - and the evidence suggests it cannot - then this increases costs of operating in British cities compared with those elsewhere. We can see how high the estimated regulatory tax is in British cities compared with their continental counterparts. This is a drag on the British economy and is a particular problem for London and the south of England; but the costs have been rising in most provincial cities since the mid 1990s.

The evidence is that it is not just mindless NIMBYism or the bloody mindedness of planners. The full analysis shows conclusively how moving from a local (or in fact partly local) tax on commercial property to a transparently national one made local communities more reluctant to accept commercial development. Who can blame them when their voters were having to pay out of their pockets to allow it?

Estimating the extra costs this created for business, because of the ultimate reduction in the supply of space it produced, shows these considerably exceeded any feasible local property tax - about twice the value in current terms. A change made to stop local communities discouraging enterprise by recklessly raising business taxes has had precisely the opposite effect! Even without that change, however, the British system was already highly restrictive.

As claimed at the start we find great variation in the impact of regulatory constraints on the supply of office space across countries and locations. It is not surprising to see tight constraints in historic city centres such as the West End of London or the city of Paris. The relatively flexible Belgian planning regime is well known and reflected in Brussels having the lowest value of the regulatory tax estimated in any European centre - though still higher than Manhattan.

The flexibility of supply also influences the volatility of prices and so (mechanically) measured estimates of risk attached to real estate assets. If supply is more inelastic then, for a given change in demand, there will be a correspondingly larger price change. Demand is inevitably subject to the economic cycle.

In the tightly constrained office market of London between 1986 and 2002 the measured variation in the quantity of new construction relative to the stock of existing space was only half that in the less constrained Brussels and price volatility was more than double.
Moreover regulatory regimes can change. We should realise that there is a regulatory risk associated with investment.

Classically this might be thought of as tax treatment of investments held abroad or the ability to repatriate profits. We do not think of planning regimes constructing regulatory risk. But they can change. If governments should become more aware of the extent of international variability and competitive drag of higher space costs, they could make systems more responsive to demand.

To an economist this has an attractive ring but it could be alarming for investors since a more flexible planning system would push down rents and capital values.If the evidence outlined here is to be believed, that adjustment in asset values would be considerable over the medium term as additional space came on stream; and it would be permanent.

Paul Cheshire is emeritus professor of economic geography at the London School of Economics

Barker, K. (2003) Review of Housing Supply: Securing our Future Housing Needs: Interim Report - Analysis, London: HMSO.
Barker, K. (2004) Review of Housing Supply: Final Report - Recommendations, London: HMSO.
Barker, K. (2006a) Barker Review of Land Use Planning; Interim Report - Analysis, London: HMSO.
Barker, K. (2006b) Barker Review of Land Use Planning; Final Report - Recommendations, London: HMSO.
Cheshire, P.C., and S. Sheppard (2002) ‘Welfare Economics of Land Use Regulation', Journal of Urban Economics, 52, 242-69.
Glaeser, E.L., J. Gyourko and R.E. Saks (2005) ‘Why is Manhattan so Expensive? Regulation and the Rise in Housing Prices', Journal of Law and Economics, 48 (2), 331-369.