Introduction
One of the thorniest aspects of the Cyprus problem is the issue of compensation to those property owners who, under the terms of the Annan plan, are not entitled to reinstatement of their property rights. The Annan plan has dealt with this by providing for a compulsory acquisition against a compensation value.
The plan recognises that there is no way the compensation value can be paid immediately in cash. It therefore provides for the creation of a Compensation Fund, which will acquire the properties in exchange for a firm promise to pay in the future by issuing to the sellers of the properties long term bonds guaranteed by the federal government.
The compensation value comprises the ‘current value’ of the property (at the date the Foundation Agreement comes into force) plus interest from that date until compensation bonds are actually issued some years later. The interest to be so capitalised shall be at the same rate as the coupon interest on the compensation bonds themselves.
The administrators of the Compensation Fund, i.e. an independent Property Board, will aim to sell the acquired properties prior to the maturity of the bonds and use the proceeds to pay the interest and settle the final redemption of the bonds at maturity. Properties, which remain unsold at the time the Property Board is wound up, shall be sold “at current value at the time of sale to the constituent state in which the property is situated”. Final surpluses or deficits go to the federal government.
Put simply, the arrangement is a large-scale property acquisition by a government on credit.
What are the economic implications of this arrangement?
This is an important question because this is no ordinary bond issue. It is extraordinary by virtue of its size. Crucially, government bonds are instruments that have a bearing on monetary conditions. What are the implications on interest rates, on the currency exchange rate, on inflation, on the government’s financial standing?
There is a tendency amongst politicians to recognise the significance of the economic consequences of a settlement under the Annan plan but at the same time to claim that, though important, these are secondary to the political dimension. We beg to differ. One needs to ask if this is going to be an absorbable shock or a devastating earthquake; especially at a time when by unfortunate coincidence the economy will be more vulnerable following the complete removal of exchange controls. The economics of the solution will ultimately determine the social and political harmony on the island. In the case of Cyprus, it is perhaps truer that economic growth will bring peace rather than that peace shall bring economic growth.
We would argue that the macroeconomic implications of the chosen method of compensation are significant, far-reaching and in all probability adverse. Identifying the adverse implications leads to recommendations to mitigate them.
Some facts about the compensation bonds
Two thirds of the compensation value shall be in the form of bonds with a 10-year duration and the remaining third shall have a 15-year duration. Both bonds are issued by a federal agency and the federal government guarantees their capital and interest. They carry a ‘fair’ interest rate fixed at the time of issue by reference to interest rates applicable at that time on federal government bonds of similar maturity. They are denominated in Cyprus pounds. They are backed by the Greek and Turkish Cypriot properties not reinstated. And, significantly, they are negotiable. Negotiability is a major feature of the bond. It means that the holder need not necessarily hold the bond till maturity. He is free to sell it to a willing buyer, in Cyprus or overseas, at a freely determined price.
There are two key things about the bond we do not know. First, the face value, or in other words the size of the issue. Secondly, the market price in the secondary market that will develop. Both of these have major macroeconomic implications.
Size of the bond issue
Estimates of the current value of Greek and Turkish Cypriot properties to be compensated vary. We understand that an estimate by the Planning Bureau puts this figure at £10 billion, which together with interest to be capitalised until issue, places total compensation value at around £12 billion. Commentators have argued that the amount could be much smaller since the Annan plan allows dispossessed owners to opt, instead of compensation, to lease their properties for 20 years or more, at the expiry of which the property use presumably reverts to them. It is difficult to gauge demand for such long-term leases and whether rents offered shall be sufficiently attractive to have a major impact on the figure of £12 billion. The Annan plan is also not as clear on this matter as it could have been. To put the bond size in context, one should note that Cyprus’ public debt as at the end of 2003 is estimated at £4.2 billion and is already just over the relevant Maastricht criterion of 60 per cent of GDP. A bond size of, say, £10 billion is 2.5 times the existing public debt outstanding.
Market Price
The bond is negotiable and the forces of demand and supply will, of course, determine its market price. There can be little doubt that, because of the sheer size of the bond and the fact that it is an unexpected windfall benefit to the recipient, there shall be an excess supply, putting downward pressure on price, quite apart from the issue of risk, which is a fundamental determinant of price for such debt instruments.
The main risk element of this particular bond is its perceived risk of default. Also, given that by the time of issue the Cyprus pound will be freely convertible, another dimension of risk relates to its perceived currency risk, which is in turn largely determined by the perceived inflation risk.
Risk of Default
The risk of default has two dimensions. As with any fund, it has to do with the level of ‘gearing’, and it also has to do with the ‘relative value of its assets and liabilities’.
Gearing
The compensation fund has financed its acquisition wholly with debt (i.e. the bonds). It is practically 100 per cent geared (the initial £100m equity put up by the federal government, possibly increased, according to the Annan plan, to £200m by international donors, being too small to matter). This adds significantly to risk.
Relative Value of Assets and Liabilities
The Compensation Fund’s liabilities are established on the basis of ‘current value’ whereas the value of the assets that support this liability is their ‘realisable value’. It is evident that current value and realisable value are not necessarily the same and may well diverge substantially from each other.
To the extent that perceived current value exceeds perceived realisable value, the Compensation Fund will have a perceived shortfall – a high risk of default. And conversely, to the extent that perceived realisable value exceeds perceived current value the Compensation Fund will have a perceived surplus – a low risk of default.
Current Value
So what is ‘current value’? The definition of ‘current value’ as defined in the Annan plan is vague and open to interpretation. The Annan plan itself admits that “expert advice shall be sought from quantity surveyors, economists and/or specialists in property valuation on the final formulation of provisions relating to assessment of value.”
The Annan Plan does set a framework for ‘current value’ by defining it as “the value of a property at time of dispossession, plus an adjustment to reflect appreciation based among other things on increase in average sale prices of properties in Cyprus in comparable locations in the intervening period up to the date of entry into force of the Foundation Agreement (based on the hypothesis that events between 1963 and 1974 had not taken place, i.e. not take into account depreciation in values due to those events; it should if possible therefore be based on comparable locations where property prices were not negatively affected by those events)”.
This definition raises a number of questions:
What are the ‘other things’ referred to on which appreciation is to be based?
What is the meaning of “properties in Cyprus in comparable locations”? Will the value of a coastal plot in Kyrenia be established by reference to similar coastal plots in Kyrenia or, say, in Paphos? The Cyprus property market in the south has had a markedly different history from that in the north.
What is the meaning of the assumption that “events had not taken place”? The history of the Cyprus property market not just in the north but also in the south has been directly related to the events of 1963 and 1974. It is safe to assume that property values in the south would have been lower, and in the north higher, had the events not taken place. And, another fine point. What is the significance of the ‘i.e.’ rather than an ‘e.g.’ in the above definition of current value? Does it imply that appreciation due to those events shall be taken into account?
We are clearly in the realm of hypothesizing. However, the prevalent view is that the current value for a Kyrenia plot will be determined by reference to actual market appreciation for a Paphos or Ayia Napa plot (without discounting the positive impact of the events of 1974 on the values of the latter). If that is the case then there is clearly a high risk that the Compensation Fund will seriously overpay.
Realisable Value
Turning to ‘realisable value’, it is evident that the restrictions that the Annan plan places on property ownership (the quotas) will have a significant depressive effect on the value of the Compensation Fund’s assets located in the Turkish Cypriot Constituent State and prevent it from achieving their true free market values. The restrictions effectively prevent the people who most desire and can afford to purchase the fund’s assets (the Greek Cypriots) from entering the market as buyers. At the same time, the seller (the Compensation Fund) is under pressure to generate from disposals, for the annual interest cost alone, sums of the order of £500-£600 million. The restricted market in the north, even with foreign interest, clearly cannot absorb such supply (property transactions in the much larger and open market of the south reached £720 million in 2002), a point already well made by Mr Constantinos Lordos. Therefore worries expressed about realisable value are well founded.
With current value likely to exceed realisable value, in other words liabilities exceed the assets, this is a fund with a negative net asset position. This, together with the gearing risk, leads to perceived high risk of default, at least as far as the Compensation Fund itself is concerned.
Of course, that is not the end of the story because there is a significant mitigating factor. The bond is guaranteed by the federal government. Will there at least be a perceived high probability that the federal government will be able to meet its guarantee obligation and bail it out? This is a major question and cannot be addressed without a forward analysis of the financial standing of the federal state, which is outside the limited scope of this article. However, we find it difficult to believe that perceptions on the size of the likely shortfall and the federal government’s creditworthiness (bearing in mind the island’s heavy financial burden for the reconstruction) shall be so favourable as to eliminate the perceived overall risk of default. We therefore conclude that such perceptions will have a significant depressive effect on price.
Currency risk
The imminent removal of restrictions on capital movements means that Cypriots will soon start thinking of their wealth not only in Cyprus pounds but also in terms of euros. Their perceived currency risk will have an important bearing on their demanded risk premium for bonds denominated in Cyprus pounds. This already applies to international players who may have an interest in the secondary market. At the very best, we would argue that the Cyprus pound is likely to have a fairly bumpy ride in the post-solution period and its value expectations will largely be determined by the extent of political stability, the record of economic growth and macroeconomic conditions in general. Hence, it is reasonable to expect a currency risk impact on the market value of the bonds.
Inflation risk
The bond is negotiable and it is safe to assume that a significant proportion of initial holders will desire to cash in. To the extent that their counter parties are either Cypriot banks or international banks with the proceeds repatriated to Cyprus, we can expect to see a sharp increase in local deposits. It is difficult to reach a conclusion other than that the increase in the money supply will cause a potentially significant rise in inflation. Given the inverse relationship between inflation expectations and the value of long-term fixed rate bonds there is also an issue of inflation risk impact on the market value for the bonds.
Summing Up
Excess supply and risk premiums clearly suggest that the bond will trade at a discount to face value. On the 10-year bond, a discount of 10 per cent implies a yield to maturity of 6.4 per cent. A 50 per cent discount implies a yield of 15 per cent.
The yield represents the return that one can earn buying or holding long-term Cyprus government bonds. Bond yields are important because they really are another name for long-term interest rates. Higher bond yields mean higher interest rates throughout the economy. Why should one hold an existing government bond that yields 5 per cent and not switch it to a compensation bond that yields more? Markets, as we well know, are very adept at eliminating such inconsistencies.
This higher interest rates outlook arises at the very time that Cyprus least needs it. It would be unfortunate to have to face prohibitively high interest costs at a time of such great financing needs for the reconstruction.
It is clear that the macroeconomic risks (interest rates, currency, inflation) and federal government risks (disproportionately high guarantee obligations) associated with the 100 per cent bonds solution are significant, far-reaching, in all probability adverse and come at a time when by unfortunate coincidence the economy will be more vulnerable following the complete removal of exchange controls. This potent mixture of risk is tantamount to playing with fire and wisdom dictates that it is best avoided. We do not believe that refinements that are limited to extending bond durations come anywhere near the necessary risk minimisation.
An all-equity alternative
An idea that has recently received publicity is that of making the compensation fund an equity fund rather than a bond fund i.e. dispossessed owners receive shares rather than bonds.
We believe that this idea has significant merits and is in the right direction. There is no risk of default, since the fund will pay out to its shareholders whatever it has collected and thus no risk on the federal government. Macroeconomic risks are also minimised because shares are not money instruments in the sense that government bonds are and have no similar direct bearing on macroeconomic conditions. Moreover, it appears to have the merit of fairness, because the ultimate value that dispossessed owners will receive is the value that these properties actually achieve in the market, no more no less. And, by making the shares transferable, risk averse or short-horizon holders should be able to cash-in and pass on the risk-reward trade-off of holding to maturity to others who have different risk profiles and investment horizons.
However, we would argue against a 100 per cent equity fund. The fact that the underlying property assets would have to be disposed in a market of limited liquidity and purchasing power and afflicted by restrictions means that true value will not be achieved. The uncertainty as to the value of the shares would make it difficult for the holders of shares to realise them at a reasonable value, especially in the initial years. Moreover, the issue of shares would be massive (five times the capitalisation of the CSE) in a market plagued by lack of liquidity, raising serious doubt as to the marketability of the shares. Shares may receive dividends at a later stage but lack the certainty of a steady income stream that is provided by interest on a guaranteed bond. And crucially, given the public’s mistrust of shares in the aftermath of the 1999 stock market bubble, it is difficult to see much public support for an all equity alternative which essentially places the risk-return trade-off wholly on the shoulders of the dispossessed owners; particularly as all discussion to date has been about offering guaranteed bonds.
RECOMMENDATIONS
A 100 per cent bond fund and a 100 per cent equity fund are the two extreme ends of the spectrum. We would recommend a combined equity and bond fund, with a predominant equity element.
We believe that the downsizing of the bond element reduces significantly the macroeconomic risks. Inflationary pressures and the upward pressure on interest rates are seriously curtailed.
The risk of default or major deficit on the federal government is also minimised, if not eliminated. For example, under a ‘1/3 bonds-2/3 equity’ scenario, realisable value need only be 33 per cent of current value for assets to cover liabilities.
At the same time, the risk-return trade-off amongst the affected parties (i.e. the dispossessed owners and the federal government) is shared between them in a reasonably balanced way. The downsizing of the bond issue together with the now low risk of default would mean that those wishing to liquidate the bonds would get a better value. It is true that the priority on liquidation enjoyed by bonds over shares would tend to keep the value of the shares low. But dispossessed owners are getting a package. Because the package includes a bond element, shareholders would afford to be patient and, after all, because of the equity element any surplus value (which could be significant if things turn out better than expected) would accrue to them.
We would recommend the application of a single predetermined appreciation factor to all properties.
A major advantage of the proposed combined solution is that, by introducing an equity element, we can also get away from the enormous, time-consuming and highly contentious task of establishing individual appreciation factors on a property-by-property-basis.
The dispossessed owners could be offered what might be termed ‘reasonable value’ e.g. the 1974 value of their property uplifted by a single pre-determined factor established by reference to the overall actual appreciation experienced in the market since then. The factor shall be the same for all and should be known at the outset i.e. incorporated in the Annan plan. Dispossessed owners will know, right at the beginning, what they are getting rather than wait for years for the Property Board to establish values. And the securities, bonds or shares, can be issued much earlier. It is true that the single factor method implicitly assumes that 1974 relative values between different properties remain unaltered. While admittedly this may introduce some distortions, we believe that the saving in terms of time and cost involved in establishing current value (which given its subjectivity is not itself distortion-free) justifies adopting this approach. Once ‘reasonable value’ is estimated, the owner would receive its nominal value in the predetermined proportion. The existence of the equity element ensures that the dispossessed owners would not lose out as a result of adopting this approach, since ultimate surpluses are theirs by right as shareholders, something that cannot be said for the all bond alternative.
We would recommend that the bond interest rate be established at the outset and incorporated in the Annan plan.
We are concerned with the provision in the Annan plan that provides that the bond interest rate shall be established at the time of issue by reference to interest rates applicable at that time on federal government bonds of similar maturity. Time of issue is a few years away. This is an open-ended commitment and is best avoided from the point of view of the federal government.
The existence of the equity element enables the adoption of a rate at the outset to be incorporated in the Annan plan without compromising fairness. We would recommend 4 per cent per annum.
We would recommend that both bond duration and the life of the fund be extended.
This would reduce the pressure on the Property Board to sell under forced sale conditions. We would suggest dividing the bond element into three equal tranches with 10-15-20 year durations. Dispossessed owners could have the option of opting for more 20-year bonds, which would carry, by way of incentive, a 20 per cent premium at maturity. The Fund would have another 10 years i.e. a total of 30, to dispose of all the properties left after the bonds are redeemed and distribute the proceeds to shareholders.
We would recommend that provisions be included that ensure the definitive resolution of any residual properties issue.
We feel uncomfortable with the provision in the Annan plan that provides that properties that remain unsold at the time the Property Board is wound up shall be sold at current value at the time of sale (does it mean current value as defined in the Annan plan or the going market price?) to the constituent state in which the property is situated. The plan is silent on what happens if the constituent state is simply unable to pay for them. Clearly, this is a valid concern and needs to be resolved at the outset. We would recommend that should the constituent state in which the property is situated fail to meet its obligation within a 12-month period then the properties shall be placed on auction and the residents of the other constituent state shall be free to participate.
