David Isaac and Neil Woodroffe
The assets of property investment companies are significantly undervalued on the Stock Exchange if the price of the shares is compared with the property valuation of the assets held. This means that the total market value of the share capital is less than the estimated value of the company’s property holdings. The estimated value is calculated here as the net asset value, which is the sum of all assets less liabilities. The difference between the market price per share and the net asset value per share — the discount on net asset value — is referred to in this article as “the discount”. It arises because investors and their professional advisers value property companies on their asset value rather than on capitalised profit, the more common approach to company valuation.
The importance of the discount is that in discounting property asset values it discriminates against property as compared with other assets, such as those held in non-property shares. It also discourages the growth of property companies through equity expansion, thus forcing them into borrowing to expand so that they become highly geared. High gearing — a high ratio of borrowed capital to equity capital — raises problems which are discused later in this article.
As well as the problems of gearing, the discount has another major effect on the growth of property companies in that it discourages takeovers and rationalisation in the sector: takeover situations raise share prices of the company being taken over and thus narrow the discount. Subsequently, however, the share price will tend to fall toward the previous discount level, the fall depending on the view that the market takes. Thus the problem is that the takeover will generally mean a dilution of net asset value per share (a greater discount) unless assets are being acquired which stand at a greater discount to net asset value. In a nutshell, if you take over a relatively strong company you could end up with a higher discount.
The discount, because it prevents full funding of assets, may also have effects on the unitised property market in that if unit prices reflect share prices then it suggests that — depending on the cause of the discount (these factors are discussed later) — the unit prices may discount the valuation of the property assets held.
It is essential, accordingly, to understand the causes of the stock market’s approach to property companies as indicated by the discount, and if possible indicate how its effects on raising finance in the property market might be minimised.
The property companies
Property companies have been a significant feature of the property market for most of the post-war period. The development boom of the late 1940s and early 1950s, based on rental growth, low building costs relative to value and low interest rates, enabled property companies to increase their equity interests in property without committing their own internal funds. Thus the original growth in property companies was financed almost entirely by borrowed moneys. A ready source of finance was found in insurance companies and pension funds as the institutions increasingly dominated the savings market.
Initially, the funds provided fixed interest finance through mortgages and debentures. Subsequently, as the funds recognised the benefits of participation in the rental growth of property, they began to engage in sale and leaseback arrangements, partnerships and direct development. These developments meant that by the late 1970s the financial institutions came to be the dominant influence on the commercial property markets through their financial control. The tax status of the funds gave them a significant advantage in direct development over the property companies.
In recent years there has been a decline in institutional interest in property investment and in direct development as the rewards from other investments have risen and the rate of inflation has abated. This has led to increased opportunities for property companies to become more innovative in raising funds, and attention being focused on corporate funding as a key source of property finance. Fixed-interest securities and discounted bond issues are recent examples of these developments. The problem of raising finance on the stock market is that the property valuation of the assets is discounted by the market. The share prices of property investment companies tend to stand at an average discount of approximately 20% to the net asset values of the companies (they were at this level at the time of the original research for this article but have narrowed recently).
The discount dissuades property companies from raising money in terms of equity and thus encourages them to borrow money and increase their gearing. This is because if the company buys £100 of assets in the market at that property valuation, the stock market will only finance them to the value of, say, £80. High gearing — a high percentage of borrowed funds in relation to equity capital — is advantageous during periods of rapid rental growth as rental profits can outstrip interest payments, to the benefit of shareholders. However, high debt payments enhance the chances of insolvency during periods of declining profitability. Thus high gearing often tends to encourage investors to regard property companies as more risky investments.
The discount on net asset value
There are three reasons for the discount:
1. problems of loss on forced sale;
2. tax liabilities; and
3. the stock market’s view of property valuation.
(1) Problems of loss on forced sale
There would be a problem of loss on forced sale if large amounts of property assets were to be forced on to the market at any particular time. This could lead to a fall in prices because of a short-term oversupply and thus the full asset value would not be realised. Therefore, the discount may be explained by investors in property company shares looking to value the underlying assets at forced sale rather than open market valuation.
(2) Tax liabilities
These arise from the specific tax position of property companies in relation to investors who hold property direct. The tax reasons thus relate to the problems of the corporate structure being an inefficient tax vehicle. This situation has been widely quoted and stems from the double taxation of income from property investment and the subsequent taxing of income from dividends to the individual investor. Second, it may be that capital gains tax gives rise to the problem of the discount, as property investment companies (unlike most commercial companies which manage to obtain significant roll-over relief) have a liability to capital gains tax on disposal of their property assets. Whether or not capital gains tax is an important factor in the discount does depend on the disposal programme of the company, and given the fact that we are dealing with investment companies in this analysis, one would only expect a proportion of each company’s assets to be disposed of in any particular year.
(3) The stock market’s view of property valuation
The third reason for the discount is thought to be the scepticism that the Stock Exchange has about property valuation methods which is evidenced by the Barkshire report’s statement on the unitised property market:
There is concern that the valuation of some buildings is uncertain because it is difficult or even impossible to test their value in the market place or to measure it easily against similar buildings.
The effect of this factor on the discount on the property valuation is similar to that of the forced sale argument quoted above, but the essential difference is that here the problem is one of overvalue rather than oversupply which gives rise to the perceived depression of disposal price.
The level of discount
Given the increasing importance of the stock market to property finance it was important to identify the level of discount and try to test the validity of the explanations given for the discount. By using a sample of property investment companies it was possible to test the market price of the share on any particular day with the net asset value shown in the last accounts. This was not a satisfactory approach, because present prices were being compared with historic net asset values. By the use of stockbrokers’ reports, however, it was possible to assess what they thought the present net asset value was, given revaluation of assets and changes in the portfolio. Thus market share prices and asset values could be tied into the same time period. By excluding two investment companies (Stockley and Rosehaugh) who, at the time of the research, because of particular circumstances in the market, were trading at a substantial premium, the snapshot provided an average discount of 18.88% on the full sample based on the analysis of historic asset values (ie the values in the last accounts) and 23.61% when viewed against prospective asset values (an analysis using data which updated the value of property assets in the company accounts).
A selection of companies studied is shown in Table 1. These average discount figures are close to 20%, which is often suggested as the arbitrarily determined discount for property companies, generally reflecting the market’s view of the importance of the factors quoted in the previous section. The research for this section was carried out in the summer of 1986 and there is evidence of a recent narrowing of discounts showing recent short-term confidence.
Testing the possible causes of the discount
Using the sample above it was also possible to test the movement of the discount against other variables and over time. The table shows significant differences between discounts for the different companies. If the market determines the discount on the basis of performance of individual companies then it should be possible to analyse performance indicators against the level of discount to ascertain whether there is any correlation between these two factors. It should also be possible to test whether the discount is related to other characteristics of property companies — for instance, size — since if the asset base is large, the market may view that company as a less risky investment.
Finally, the attractiveness of the portfolio of properties held should also affect discount. Although portfolio analysis is a difficult area to deal with it should be possible to determine, for instance, a predominance in either retail, office or industrial property. If investors, as perceived by yield indications, are more likely to wish to invest in retail rather than office and office rather than industrial property, then an analysis should give some relationship between discount and the nature of the portfolio, in the sense that unattractive types of property should result in higher discounts. Although this analysis is fraught with difficulties in consideration of whether properties are prime or not, and as to the design characteristics and locations of properties, it was felt that some initial testing could be justified.
Our research therefore also tested the level of discount against company performance, using return on capital employed and the price: earnings ratio; stability, using firm-size as measured by net asset value; and, finally, the type of portfolio.
Analysis showed that no relationship existed between the level of discount and any of these variables. Although there were significant variations in the level of discount between companies they were unrelated to the variables identified above. Such variations are also inconsistent with the view that the market imposes an arbitrary 20% on all property companies, but the variables causing the variations are difficult to identify.
Further study looked at shares in other sectors of the market, but the problem here is that shares other than in property investment companies are valued on an earnings rather than an asset basis. An earnings approach entails taking the historic profits and applying a price: earnings ratio to them or, alternatively, using a discounted cash flow approach to find the present value of an estimated future stream of earnings. The problem of the discount does not apply to companies outside the property investment sector because in these cases investment returns are derived from income and profits, not based on capital appreciation on revaluation of the assets held by the company.
Finally, changes in the level of discount over time were measured against changes in property yields and taxes. This appeared useful in two respects. First, if the level of discount represents the investor’s view of the quality of the investment, then there should be some relationship between changes in yields and the level of the discount (ie a fall in yields should lead to a reduction in the discount). Second, if the tax position (corporation tax and capital gains tax) is an important factor than a change in corporation tax rates and the improved indexation of allowances against capital gains should reflect changes in the level of discount, as the tax inefficiency of the company is reduced. However, neither the successive reductions in corporation tax from 52% to 35% between 1983 and 1987, nor improved indexation provisions on purchase costs in relation to capital gains tax assessment, appear to have had a systematic impact on the discount. This suggests that while tax efficiency improvements are necessary for property companies to make the best of their opportunities they do not appear to be the prime cause of the discount.
The relationship with yields showed the discount rate to be more volatile than yield and to be fluctuating in a cyclical way to return to an average discount of around 20% regardless of yield. Thus a yield change, making a difference in the attractiveness of property, will encourage a short-term change in discount, the discount varying directly with yields. In the long run, however, there is a tendency for the average discount to move back toward the 20% level even though yields may remain static. Thus the discount is a less sure way of measuring confidence in property investment as determined by yield.
Conclusion
The period of institutional dominance in the property market gave the impression that the assets of the property companies would in time be swallowed up by the institutions. The fading performance of property has led to less activity in the market from the institutions and the need to raise alternative funds from corporate sources to replace institutional funds. The major problem with this alternative form of funding is that the stock market discounts the valuations of assets of property investment companies. The size of discount does not appear to be related to factors linked to market expectations of future income and asset growth as evidenced by the analysis shown here, although one would have expected it to be the case. Nor does it appear to be the case that the specific tax position of property companies is the key determinant.
Therefore, this initial analysis suggests that further research is necessary to identify the disparities between valuation methods adopted by the property profession and the attitudes of stockbrokers and investment analysts. Such a reconciliation is necessary to improve the flow of funds into property and assist the investment status of property assets.
