Recent securitization in real estate financial markets has changed the way in which the traditional three-phased real estate cycle will operate in the future. This column explores the nature and effects of those changes.
The real estate cycle contains three phases, as often explained in this column. The development boom phase starts at the peak of a general business cycle expansion. But when a lot of properties started in this boom come onto the market, the general economy has gone into a recession, so there is a space surplus on most markets. This causes the overbuilt phase. But ultimately the general economy starts to expand again, and demands for space begin rising. This gradually soaks up existing space, causing vacancies to fall, rents to rise, and property values to rise, although little new construction occurs. This is the gradual absorption phase. Eventually it gives way to a new development boom.
The three phases of the real estate cycle
* Development boom phase
* Overbuilt phase
* Gradual absorption phase
During 1998, we were in - and in 1999 we are still in - a general development boom in most U.S. commercial property markets. New construction, both build-to-suit and speculative, sprang up everywhere. Demands for space have been growing too, often faster than new supply. But the amount of new supply in the pipeline and planned for 1999 will probably exceed increases in space demand because those increases are slowing down.
When 1998 began, we were also in what might be called a "bull market" for existing properties. REITs were bidding aggressively for such properties to expand in size. REITs had been able to raise capital easily because stock prices were rising so strongly, so they could outbid traditional investors. And REITs were under tremendous pressure to grow larger so they could attract large-scale institutional investors who want to deal only with highly liquid stocks - and "normal" liquidity rises along with total size.
However, I pointed out then that when the stock market stopped rising so fast, REITs would be unable to float secondary issues and raise capital so easily. Three events proved me right. The first event was that REIT stock prices did not rise along with the rest of the market in the first half of 1998. The second event was that the stock market, itself, took a nosedive in August as a result of a financial panic. This cut off the supply of new capital to REITs almost totally, causing many to back off from aggressive acquisitions.
The third event was a paralysis in the CMBS market caused by the sudden appearance of "accumulation risk." That is the risk that a conduit lender faces when it is issuing and accumulating a whole lot of new loans in order to later issue a very large bond issue based on those loans. During that accumulation period, many conduits borrowed short to finance their long-term mortgage loans. But borrowing short to finance long-term loans is inherently foolish because market conditions can suddenly change, as they did. The interest rates the conduits had to pay shot upward, but the prices they could get by selling their bonds fell sharply - leaving many in a loss position.
Moreover, there was a universal flight from high-risk ventures caused by the global panic. So few investors would buy the B-pieces of the CMBS bond issues - the highest-risk pieces. This left conduit issuers with huge amounts of unsold B-pieces, causing immense losses for many. Yet conduits had been financing a lot of the proposed new projects forming the new development boom. So when funding from conduits dried up overnight, that put the brakes on a lot of new development.
When the REITs withdrew from aggressive property buying, a major cause of the preceding run-up in property prices disappeared. So prices in those markets began to recede, though not rapidly at first. Therefore, around mid-1998, we passed the peak in real property prices in this real estate cycle and this general business cycle. From here on, prices will be moving downward or sideways in most markets. They have already fallen 10% to 30% in many cases.
It might be argued that prices could be driven up again if REITs come back into these markets as aggressively as before. But this is unlikely. One reason is that REIT stock values have not risen along with the S&P 500 and are still well below their highs - by degrees varying with types of properties. Yet I believe that the total real estate credit freeze caused by the fall panic will ease by this spring - if we do not have further market deterioration in global financial conditions.
But REITs will still not be able to fund as aggressive acquisitions as they were in early 1998 because the cost of capital to them will be too high. Many investors realize the real estate value cycle has passed its peak, and that widespread new development is creating more competition. So they do not believe REITs can grow incomes as fast as in the past few years.
A second key development has been a slowdown in new project financing caused by both the REIT withdrawal and by the drying up of CMBS financing. Many projects planned for 1999 and beyond have been postponed - perhaps indefinitely. This will slow the overbuilding that normally occurs in a development boom.
Promoters of securitized financing have been saying, "We told you that overbuilding would be stopped by the brilliant foresight of public market investors, who have choked off new credit to REITs because they saw overbuilding coming." There is some truth in this claim, as shown by the fact that REIT stock prices have not risen along with other stock prices.
But the principal reason credit has been cut off to new development was not the failure of investors to support REITs, but the crisis in CMBS markets. They were supplying most financing for new development, which was mainly carried out by traditional developers using CMBS and other mortgage financing, rather than by REITs.
Thus, the major reason for cutting off credit from new development was the worldwide flight from risk that caused all investors to demand higher yields from any type of risky projects, such as new real estate projects. This did not indicate any superior foresight by real estate investors in securitized forms.
Is the new development boom over because there will be inadequate financing from here on? I believe some mortgage financing will reappear this spring and enable a resumption of many new development projects. However, that will happen only if (1) there is no new global financial panic and (2) the U.S. economy continues to grow by at least a moderate rate, thereby continuing the increase in space demand.
This mortgage financing will not only come from some reactivated conduits, but also from traditional mortgage lenders who see high yields on mortgages compared to other fixed-rate securities. But there will be less overbuilding in 1999 than would have occurred without the recent financial panic and, therefore, no general oversupply in real property markets. This may cause a much shorter overbuilt period in this cycle.
This analysis reveals that the increased importance of public markets in providing capital for real estate has modified the three-phase real estate cycle to some degree. The main change has been an increase in the volatility with which capital will flow into real property markets. Since public markets have been providing a lot of that capital through both REITs and CMBS issues, variations in the availability of capital flows from public markets will affect the pace at which properties can be both bought and developed. But that pace will depend upon what happens to REIT stock prices.
During periods when REIT share prices are rapidly rising, property prices in private markets can be driven upward faster than in the past. Then REITs can raise a lot of capital in public markets to use in buying properties in private markets, accelerating the normal cyclical increase in prices there. This can shorten the gradual absorption phase by driving prices up enough to stimulate new construction.
But stock market values experience sharp downward movements much more often than private real property prices. Stocks fell sharply in 1987 and 1998 when the overall economy, and private property prices, did not fall. Yet REIT share prices are highly correlated with stock prices in general during broad general price declines, as this most recent credit crunch clearly proves. So REIT share prices fall more often than private property prices.
When REIT share prices fall with stocks in general, REITs are less able to float new stock issues - so they cannot get more capital. This slows the total flow of capital into private property markets much more than when REITs were not such an important source of capital there. That interruption can affect the general level of existing property prices - reducing them or flattening them - and affect the pace of new development - slowing it. So the development boom phase may no longer proceed uninterrupted from gradual absorption to overbuilding. How long this interruption will last depends upon what happens to stock prices and to REIT share prices in particular. The former have risen again, but the latter have not.
Thus, the increased importance of public markets can affect the way the real estate development cycle behaves, even if the economy in general keeps on expanding.