The end of boom and bust – sounds too good to be true?

The challenge is to curtail excessive commercial property lending at the end of the cycle to reduce the risks to the stability of the UK financial system. While this may sound very boring, getting this right is more likely to affect the person on the street than any other single action the property industry ever takes.

Enthusiastic end-of-cycle property lending is effectively ‘betting the bank’. The long-term value approach in question is intended to inform lenders (and regulators) when property is becoming overvalued and to make it easier for them to take steps to moderate or curtail their commercial property lending two or three years before the boom turns into a major downturn.

The PIA looked at a range of long-term valuation methodologies and through back testing identified adjusted market value (AMV) as the one that worked best. AMV compares current values with the market long-term trend (adjusted for inflation), a concept that has been around for a long time.

Back testing establishes correlations, including showing how historically the predicted overvaluation has translated into falls in capital values over the five years following the valuation date.

Identifying correlations is the game-changer. Lenders and regulators using AMV will not only be aware that the market is overvalued, but will also have a reasonable idea of the probability - and likely severity - of the potential fall.

Not an exact crystal ball

While this sounds like an exacting science where lenders and regulators can look at formulae to avoid losses, unfortunately it is not quite that simple.

There is the usual health warning that “past performance does not necessarily guarantee future results”. AMV has worked historically, correctly anticipating each of the four major crashes in the UK market in the past 100 years, but this does not mean that it will identify the next crash with pinpoint accuracy.

It is difficult to predict the exact timing of a fall, which may happen at any time in the five-year window (or not). If AMV was that accurate, the whole market could use the methodology to time their buying and selling and clean up. Also, lenders should recognise that while expected average falls in market values may be at acceptable levels, individual asset falls may be significantly greater.

However, the risks of not adopting a long-term value approach far outweigh the limitations.Historically, the bigger the likely crash, the more accurate AMV becomes - at precisely the time it is needed. At 15% overvalue, the probability of a (real value) fall of 30% or more has been 88% (96% for a 25%-plus fall).

At this point, any responsible lender/regulator should recognise that the risk/return dynamics of lending are looking increasingly toxic, particularly when lending 65% to 70% or more of market value.

Anyone querying the intellectual validity of using AMV, a simple long-term value average, should ask themselves this: will using AMV significantly reduce the risks of excessive lending at the end of a major cycle, when compared with carrying on using just market value? Of course it will - dramatically.

All this sounds promising but there is one rather big caveat: this will only work if lenders and regulators hardwire long-term value methodologies into the centre of their risk management systems - and then do not flinch from taking action at the appropriate time, in spite of increasing pressures that will encourage them to do otherwise. That’s the next major challenge.

Rupert Clarke is chairman of the PIA Long-term Value working group