|
Article Excerpt
This study examines the correlation of the frequency of the sale of investment-grade property with national, regional and local variables, including property-and owner-specific characteristics. More specifically, the study identifies the primary factors that "explain" intertemporal changes in transaction frequency in the underlying properties used to develop the National Council of Real Estate Investment Fiduciaries (NCREIF) index. Understanding these factors yields important information that can be used by firms when forming market expectations and developing investment strategies.
**********
Real estate market transaction activity is closely monitored by private sector investors, analysts and advisors, as well as by public sector participants. Significant increases or decreases in the level of prices or the frequency of transactions are often reported with some fanfare in the popular press. Understanding the factors affecting real estate price levels and their movements has been the focus of much research and analysis: however, relatively little attention has been given to examining transaction frequencies. In private real estate markets, changes in the frequency of transactions, as measured by changes in the number of properties sold from a stock of properties comparing one period to another (e.g., month to month, year to year), have been shown to be related to asset price movements. (1) Thus, transaction frequency is a key indicator of current conditions of the real estate market. In addition, it is widely viewed as an important "leading" indicator of the expected general conditions of local, regional and national economies. By monitoring transaction levels, market participants are better able to form market expectations, evaluate prices and develop investment strategies. This study examines the underlying factors that affect the probability of property sales occurring from period to period and therefore the transaction frequency. Identifying and evaluating the relative effects of these factors is fundamental to the analysis of the real estate market and its submarkets.
To date, research on the frequency of real estate transactions has focused primarily on the sale of existing residential properties and the demand for new residential construction (e.g., important works include Jaffee and Rosen 1979, Thom 1985, Stein 1995 and Rady and Ortalo-Magne 2001). (2) We know surprisingly little about the private market "transaction cycles" of investment-grade commercial property, despite its importance to the institutional investment community. Even less is known about the factors underlying these sales cycles and whether the relationship of "concurrent" or "leading indicators" of sales activity can be properly characterized.
This study identifies the relative correlation of national, regional and local variables, including owner- and property-specific variables, with the likelihood of investment-grade property sales activity. Specifically, we examine factors that "explain" intertemporal changes in sales volume in the underlying properties used to develop the National Council of Real Estate Investment Fiduciaries property index (NPI). In addition, the study explores whether the relative importance of these factors varies across property types. Our primary focus is on intertemporal sales variation, but the research question also is relevant to cross-sectional studies.
The study is organized as follows. In the next section we discuss the interaction of buyer and seller reservation prices resulting in intertemporal changes in liquidity and the conditions leading to procyclic behavior in the transaction cycle. An econometric model describing the primary factors affecting transaction probability is presented in the third section. The data are described and empirical results are reported in the last two sections.
Transaction Frequency, Market Liquidity and the Interaction of Reservation Prices
Though related, it is important to distinguish between the concepts of transaction frequency and market liquidity. Transaction frequency refers to the number of transactions that occur in a particular market during a particular period of time (e.g., 120 properties per year). The number of transactions is generally affected by market conditions (including the size of the market), property conditions and other factors influencing buyer and seller decisions. Transaction frequency is easily observed and represents a key indicator of the liquidity in the market.
Market liquidity refers to the ease, or speed, at which properties transact or are expected to transact. One measure of the liquidity of a market is the reciprocal of the transaction frequency. For example, a market with an average transaction frequency of 120 properties per year constitutes a transaction "speed" of about 3.0 days per transaction (365/120). (3) As with transaction frequency, comparison of this measure across markets requires controlling for the relative size of each market.
More commonly, market liquidity is viewed as the expected time required for a particular property to transact (e.g., 30 days). Using the example from the previous paragraph, if there are 10 sellers (properties) to each buyer, then the expected market transaction rate on each (constant-quality) property extends to 30 days. Thus, market liquidity depends on the relative number of buyers and sellers "in" the market at a particular time--reflecting the conditions of the market, property and other factors affecting their purchase/sale decision. It is important to note that the relative change in the number of sellers and buyers is fundamental to changes in both market liquidity and transaction frequency. If market size (i.e., the number of buyers and sellers) is held constant, transaction frequency and market liquidity are directly related, and movements in both reflect the changes in conditions affecting buyer and seller decisions.
Two stylized facts have been observed with respect to the transaction frequency (and liquidity) of private real estate markets. (1) Transaction frequency varies dramatically from period to period and market to market. Property owners can sell more assets or sell any given asset quicker and easier (holding price constant) when there are more buyers in the market (i.e., the market is more "liquid"). Alternatively, property owners can sell the same number of assets (or a given asset in the same amount of time) at higher prices, ceterus paribus. (2) Transaction frequency (and liquidity) is positively correlated with the asset market "cycle." Controlling for market size, transaction frequency is typically greater when the property prices are relatively high and/or rising, and it is lower when prices are relatively low and/or falling. Relative to the general economic conditions, changes in the frequency of transactions are typically found to be pro-cyclical and persistent. During "up" markets, capital flows into the real estate sector, there is a greater volume of trading and it is "easier" to sell assets. Just the opposite typically occurs in "down" markets.
[FIGURE 1 OMITTED]
The stylized relationship between transaction frequency and property appreciation is shown in Figure 1. The annual appreciation rate of the capital component of the NPI is denoted by the solid line for the period 1984 to 2001. The annual appreciation rate is charted along with the percentage of properties in the National Council of Real Estate Investment Fiduciaries (NCREIF) portfolio (denoted by gray bars) that transacted each year. A strong positive correlation between periodic movements in the annual transaction frequency and the rate of appreciation can easily be noted. During the economic downturn of the early 1990s, both the percentage of transactions and the annual rates of appreciation experienced persistent declines from 1990 to 1992. Transaction frequency and appreciation rates then rose consistently until peaking in 1997 and 1998, respectively. (4)
Asset prices are affected by changes in liquidity. In markets of equal size, higher transaction frequency indicates greater market liquidity. (5) By monitoring transaction activity levels, market participants are able to assess current liquidity levels and form better market expectations with which to evaluate prices.
Changes in transaction frequency can be described by examining the relationship of the relative movements in the distributions of potential buyer and seller reservation prices. To describe this, we consider the search model presented in Fisher, Gatzlaff, Geltner and Haurin (FGGH 2003). Their diagram of the stylized distribution of the potential buyers and sellers is shown in Figure 2.
In time t of Figure 2A, the vertical axis measures the number of potential buyers and sellers, while the horizontal axis indicates the relative reservation prices, assuming constant-quality assets and heterogeneous agents (buyers and sellers). The reservation prices are defined as the prices at which potential buyers and sellers will stop negotiating or searching for a better deal and complete a transaction (FGGH 2003). The left-hand-side distribution represents the distribution of potential buyers and the right-hand side the distribution of potential sellers.
Our interest is in the overlap region where some buyers are willing to pay more than sellers are willing to accept. The number of buyers willing to transact at some price, P, on the horizontal axis is represented by the area underneath the buyer reservation price frequency distribution to the right of P. Thus, the number of potential transactions in the asset market during the period t is...
|
