What is valuation, if anything?

Getting (but not rushing!) towards the end of my career – a good deal of which has been taken up with valuation – I have been thinking increasingly about what valuation might mean in a property context. Does it, I wonder, mean much at all?

I’m not talking about the decisions of arbitrators, tribunals and courts, or the valuations evidence that is put before them. This is a precise analytical process, in which the link with the real world is often, and inevitably, tenuous. When a valuation problem is put before an arbitrator, tribunal or court, it has to find a definite solution. It often has to look at things from a rather artificial point of view in order to achieve that solution. It can’t say “It’s probably between £X and £Y”: neither the “probably” nor the range is acceptable.

Further, it’s worth making the distinction quite clear between value in the normal, linguistic sense – what the RICS call “worth” – and value in a valuer’s sense. When a valuer talks about valuation he is not talking about as it were evaluation – an assessment of intrinsic value. He’s making an estimate of what the property would sell for at a particular date. I notice that even some of the RICS documentation on valuation blurs this distinction. In this article, I’m only looking at one issue: “price payable”; not “intrinsic value”. Warren Buffett says price is what you pay; value is what you get. Property valuers are only concerned with price.

These thoughts were stimulated by three cases in which I’ve been involved recently. Obviously, I have to anonymise these, so I anonymise the price by taking a base of £100, rather than the actual price; and I anonymise the properties (and the parties) by not explaining exactly where or what they were.

Case 1

This is a building in a large provincial city which had changed hands at £100 only a few months before, and been valued by one of the big firms at £100. In the context of an unrelated dispute, I had to discuss the case with a very experienced opponent in that city. Neither of us could find an objective way of arriving at a value of greater than £60 or thereabouts.

Was its value £60 or £100? The margin of error was 40% or 67%, depending on which figure you take as the base.

Case 2

This is a building in central London. In the run-up to the sale, the agents and I had to consider what might be achieved. The vendors were advised that it was reasonable to expect £100, but perhaps that caution should be applied even to that figure. Once the process got going, a rethink suggested that the property might achieve as much as £120. My belief is that no valuer, asked to value the property in advance of actually attempting to market it, would have valued it at significantly more than this. In fact, the clients asked that the quoting price be set at £140. This produced an enormous response, which resulted in the property being sold ultimately, after various competitive bids, at £170.

Which is the value – £100, £120 or £170? The margin of error was between 20% and 70% depending on which figures you take.

Case 3

This is an industrial building in a provincial town, valued by several different valuers at around £100 two or three years running, and finally sold for something like £20.

Which is it? Margin of error a staggering 400% or 80%.

What all these examples tell me is something horribly close to the idea that valuation actually doesn’t mean anything at all. So what is it intended to mean? As I’ve mentioned in the section about Tribunals and Courts above, one thing it can mean is simply an objective assessment of the likely price payable having regard to the evidence available to support the required analysis. On the rare occasions when this process is carried out and a transaction subsequently takes place, the experience of the valuer in being startled by the figures achieved must be a pretty common one, for better or for worse. However, if the analysis purports to be no more than that, then it is valid in its own terms: it’s internally consistent. What it isn’t, however, is a prediction of what the property will in fact be sold for in the open market to anything like a meaningful tolerance.

Very often, a valuation is little more than a repetition of the price just paid; that may be the case of the provincial city mentioned above. However, a document which merely says that the price just paid for the property is its value may be correct, but is also useless. The valuer has to be able to produce some objective analysis which does not rely on the price just paid to arrive at the conclusion that that price is correct. If he can’t, the valuation isn’t worth the paper it’s written on. Some objective criteria are required other than the price paid.

One further thought is that the margins of error normally respected by the Courts and the profession may be wildly inadequate: each of the cases mentioned above seems to suggest they are. I’ve seen it suggested that the margin may be as low as 5.00% either side, and I don’t believe I’ve ever seen it suggested that it should be more than 15%. How that result can be squared with the four cases mentioned above is beyond me.

Guidance Note 5 (valuation uncertainty) of the Red Book attempts to deal with this issue, but goes nowhere near meeting the problem.