Rating agencies appear to have played a leading role in the current crisis that the real estate market in the UK finds itself in by overestimating the credit risk of property-backed debt instruments. The public should be able to expect – and to be protected by – the objective exercise of professional judgement by a responsible and regulated group, and the rating agencies failed in this regard. Did valuers also contribute to the crisis, by failing to discourage banks from lending high proportions of the ramped price of property, and by overestimating the value of property assets and property funds?
The RICS Valuation Faculty Board recently issued clarification on how valuations should be reported in times of valuation uncertainty. As the RICS states, “The valuation of … property in the current market is unusually challenging, due to a reduction in the number of comparable property transactions in what is a rapidly changing macroeconomic climate”. It approved the concept of a conditional valuation, effectively suggesting that valuation is impossible when there is no market.
In the absence of continuously traded, deep and securitised markets, commercial property valuations perform a vital function in the property market by acting as a surrogate for transaction prices. Property valuations are central to market efficiency, but within both the professional and academic communities, there is considerable scepticism about the ability of appraisals or valuations to fulfil this role in a reliable manner.
There has been a considerable amount of research into the operation of the valuation process in various parts of the world, especially in the US and the UK, dealing in particular with valuation accuracy and client influence. Given the untold damage done by credit rating agencies to the valuation profession, as defined in the broadest possible sense, valuer independence and freedom from conflicts of interest are vital to the continued operation of the real estate market. Right now, we face a hiatus and a withdrawal of liquidity. What role does valuation play in this?
The Value of Accuracy
There is a consensus that individual valuations are prone to a degree of uncertainty. At the macro level, it is clear that few analysts accept that appraisal-based indices reflect the true underlying performance of the property market. It is commonly held, for example, that such indices fail to capture the extent of market volatility and tend to lag underlying performance. As a consequence, issues such as the level and nature of valuation uncertainty and the causes and extent of index smoothing have generated a substantial research literature.
Some of this research indicates that valuations both lag the market and smooth the peaks and troughs of “real” prices. Valuations – and, if valuations affect market psychology and hence prices, real estate prices – can be “sticky”. Sticky is a term used in economics to describe a situation in which a variable is resistant to change. A variable that is sticky will be reluctant to drop even if conditions dictate that it should. The suggested reasons for this are anchoring, the tendency of a valuer to produce a valuation that is affected by last year’s valuation, temporal averaging (year-end valuations have to be undertaken over a period of up to three months) and the lagging effect of comparable evidence.
The resulting valuation “smoothing” has been widely analysed. It is generally presumed to reduce the reported volatility – or risk – of real estate investment below the real level of risk suffered by investors who have to sell in a weak market or buy in a strong one.
Valuers may be influenced by their clients. This is likely to be more damaging, as client influence creates a conflict of interest between professional responsibility and client satisfaction, interfering with objective professional judgement.
Attention has been paid to the moral hazards or conflicts of interest present when the auditing firm provides additional non-auditing services, with the result that the US Sarbanes-Oxley Act of 2002 forced the providers of auditing services to be more accountable regarding client influence. But such issues are not unique to the auditing/accountancy/consultancy profession, and there is some evidence of problems existing within the client-valuer relationship.
In the UK, a distinction is made between independent valuers, who are unable to act for the client in other transactions and activities, and external valuers, who are. This is clearly of relevance in determining the potential for conflicts of interest. Issues raised by an investigation of the “moral hazards” (conflicts of interest) involved in the principal-agent (client-valuer) relationship include incentives to provide a confirmatory valuation when fee income depends on completion of a transaction and a general drive to keep clients happy and cross-sell non-valuation services to clients. Fund managers have clear incentives to attempt to influence valuations because they are used in the measurement of performance.
What does this mean? It suggests that valuations change in response to client pressure, especially when the valuations are important, typically at the year-end in December. Research suggests that valuations may be distorted by the influence of clients, as well as by the procedures that form the valuation process. Valuations are used to re-negotiate prices (which in turn may be used by valuers to support subsequent valuations).
In addition, valuations often go through a process of negotiation with clients, and this is especially evident at the year-end December valuation. This may lead to improved valuations, bolstered by enhanced information quality, but may also lead to biased and less valid figures.
If the latter is the case, valuations may be far from independent of the market. Chart 1 below shows the December or year-end effect in the IPD Monthly Index.
The lack of independence between valuation and achieved sale prices means that a natural lag would appear to be built into monthly and quarterly indices in an upswing or a downswing. This became a huge issue for UK open-end property funds in 2007 and 2008.
Chart 1: Total Returns Based on the IPD UK Monthly Index,
Dec 2006 – Dec 2008 (%)
Building a Case
A lot of money is needed to build a diversified real estate portfolio. Investors require more significant capital investment to reduce the specific risk component of the portfolio to a desired level. For example, research suggests that more than £1 billion (€1.06 billion) is needed to build a diversified portfolio of London offices with a 2% tracking error against an index. This presents a very strong case for using an unlisted property fund focused on London offices. Assuming that such a fund is financed by 50% debt and 50% equity, 20 investors committing £25 million (€26.5 million) each will produce enough capital to achieve the diversified fund. Yet the investor’s £25 million is enough to buy only one or two London offices of average lot size.
This argument could also be used to justify investments in listed property securities. However, the pricing of listed REITs and property companies will vary from real estate prices and in the short to medium term (zero to five years) distort the performance of the securities relative to the underlying real estate market. Significant discounts and premiums to net asset value (NAV) can affect pricing for periods as long as five years.
On the other hand, core open-end unlisted funds appear to track real estate NAVs. The values – or valuations – of core (lower risk, often open-end) UK funds as measured by the IPD UK pooled property fund indices have tracked the UK IPD index of direct real estate returns. This is supportive of an investment in open-end funds but simply means that appraisal-based returns on core funds track appraisal-based returns on the index. This does not purport to show how trading prices track the index. This became a problematic issue in 2007 and 2008.
Take the case of a typical UK open-end fund in 2007. The fund is valued monthly by a well-known chartered surveying practice/property services provider. The valuation instruction is to estimate the asset value of the properties in the fund; any adjustments needed to estimate the fund NAV are usually undertaken internally as an accounting function, dealing primarily with the addition of cash holdings and the deduction of debt.
From January to June 2007, the IPD Monthly Index had been showing positive capital growth of between 0.27% and 0.46% for each of the six months of the first half of the year. Similarly, the fund showed positive NAV growth.
In the summer of 2007, there was a change in market sentiment. This was evident in some circulars from fund managers to their clients warning of poor returns to come. It was reflected to some extent in the IPD Monthly Index, which produced its last positive total return of the cycle in July 2007, disguising a tiny fall in capital value of 0.22%, followed by a negative capital return in August 2007 of 0.4%, marking, in retrospect, a turning point – but hardly indicative of a crash. The fund’s values were flat over this period, as they were through the end of September 2007.
However, the IPD Monthly Index for September 2007 took a more significant downward turn of more than 1.5%, a figure that was not available to the valuer of the fund’s properties at the time the September 2007 valuation was produced. The fund’s values stayed flat over the summer quarter, while the IPD Monthly Index fell by more than 2% for the quarter. This might easily be explained by a difference between the properties in the index and those in the fund.
Open-end fund units can be redeemed on demand, and new investors will normally be allowed and encouraged to buy new units on demand. The fund manager will issue units at NAV plus an allowance for the costs of buying new properties with the new cash (the offer price), and will undertake to return capital to the investor at the latest NAV estimate less a deduction for trading costs (the bid price). (Technically, the NAV is adjusted to offer price by adding real estate acquisition costs and offer is reduced to bid by deducting the round-trip costs of buying and selling real estate.)
Open-end funds in the UK typically have quarterly redemptions. The quarter-end NAV sets the bid and offer price. In mid-October 2007, the fund’s manager found itself in a “challenging” position. Market sentiment had clearly changed. The UK REIT market had moved to huge discounts to NAV, moving downward from a small premium in March 2007 to a discount touching 20% in September 2007. Chart 2 above shows this clearly. Meanwhile, the fund’s NAV had stayed flat over the summer.
Since late 2007, Julian Schiller’s team at Jones Lang LaSalle (JLL) has been performing a unique service for property investors, by reporting estimated premiums and discounts to NAV in the secondary market trading prices of UK unlisted closed-end funds. As a piece of research, this can be criticised as being opinion-based and partial, but it is based on secondary market deals where possible, and its value is significant because for the first time we can compare prices in the unlisted fund market with prices in the listed market.
In September 2007, unlisted funds were generally in a strange position – secondary trading had dried up as the market sought a new price level, as JLL’s estimates of closed-end fund premiums and discounts show. Only by November 2007 had the repricing become evident.
Professional investors could see a clear arbitrage opportunity and were prompted to exploit it. They served redemption notices effective at the end of September 2007, expecting to be paid out at the NAV-based bid price. This was broadly the same as it had been in June 2007, at the peak of the market, but by October 2007 values were clearly falling. The fund manager was faced with the prospect of selling properties in a very weak market for prices that would clearly not be as high as the September 2007 NAV.
This might be a fair game as far as the exiting professional fund of funds manager would view it, but it might not be fair from the perspective of an existing unit-holder who wishes to stay in the fund. If a large number of properties are sold considerably below NAV, but existing unit-holders are paid out at NAV, the remaining unit-holders suffer the loss. If this is the case, then all unit-holders will be tempted to exit at the same time, challenging the continued existence of the open-end fund, which has been a perfectly acceptable and stable investment for many small pension funds for more than 15 years since the last market crisis of 1991.
The trust deed allowed the fund manager to defer redemptions and also to make adjustments to the NAV in exceptional market circumstances. It was judged that these circumstances were indeed exceptional, and a large discount was applied to the NAV in October 2007, and a further discount was applied in January 2008; redemptions were deferred, but sales took place and exiting investors (who had been able to withdraw their redemption notices, given the downward price revision) were paid out on time, albeit at the lower bid price, by now revised down to around 80% of the NAV.
The consequences are disturbing. While unlisted fund NAVs may continue to track the IPD index and demonstrate high correlation, there is less reason to suppose that will reflect either the trading prices of funds on the secondary market or the prices received by investors exiting open-end funds. So, were valuers wrong?
Where there is little or no market evidence, valuations become difficult. REIT and fund discounts may provide a signal to the valuer, but they provide dangerously volatile signals. Nonetheless, there is strong evidence that the UK valuation profession has been more responsive than any other and, given the history of the UK market, which hit the buffers in 1973–74, 1990–91 and 2007–08, it may be that this crisis is an unavoidable, once-in-17-year event against which mere valuations cannot be expected to provide much defence. We await, with a grim sense of inevitability, the delayed evidence of a fall in certain continental European markets, already flagged by the secondary market for closed-end funds in which transactions are taking place at truly shocking discounts to NAV.
It is important to regulate the valuation process in order to protect the valuer from undue pressure, to create transparency in the process and in the outcome, and to maintain both actual and perceived valuer independence. Because of this pressure, regulation of the valuation process is increasing and will continue to do so to ensure that valuations are objective and transparent.
So did valuers also contribute to the current crisis by overestimating the value of property assets and property funds? Despite the impression created by the evidence, nothing in this opinion piece suggests that valuers did anything wrong.
As long as they can claim to have acted independently, and professionally, in line with RICS guidance, and without the clear conflict of interest suffered by other market players, especially rating agencies; then the problem of the NAV discount is in the nature of the asset class, crystallised in severe downturns, and not in valuation error.
In a healthy economy not driven solely by fear and greed, academics are right to observe and judge, the RICS is right to promote conditional valuations, and professionals need to be protected from undue market influence in the independent exercise of their professional responsibilities.
Professor Andrew Baum is chairman of Property Funds Research (PFR: www.propertyfundsresearch.com), based in Reading, UK. PFR was recently acquired by Feri EuroRating Services AG. This article is based on work undertaken at PFR and the new, third edition of Property Investment Appraisal, Andrew Baum and Neil Crosby, Blackwell, 2008.
This article appeared in the March 2009 issue of The Institutional Real Estate Letter – Europe, published by Institutional Real Estate Inc (www.irei.com).