Constructing real estate investment portfolios: how to use models and a few tools to build a diverse portfolio

Business Economics, Oct, 1999 by Timothy W. Viezer

This paper presents a series of useful models and tools for an industry that prides itself on "insider information" and local market knowledge in the selection of one investment at a time. The emphasis in this paper is on the construction and monitoring of portfolios of real estate investments.

The general principles in this paper are applicable to business economists in any industry. The specific applications have relevance to institutional investors (e.g., life insurance companies, commercial banks, thrifts, credit companies, pension funds, publicly held real estate investment trusts and real estate operating companies, private real estate investment trusts, mortgage conduits and mortgage backed securities, and both public and private real estate syndications) as well as regulatory agencies (e.g., Federal Deposit Insurance Corporation, the Office of Thrift Supervision, and the Office of the Comptroller of the Currency).

Useful Models

Sergeant Friday had it wrong: "just the facts" are not enough. Decisionmakers need useful models to give meaning to the maze of facts that would be otherwise confusing and useless and to put facts into a more usable, practical form. Unfortunately, the necessarily abstract nature of economic models often prompts some to conclude incorrectly that models are impractical and unrealistic, divorced from the facts and realities of the world. As G.E.P. Box once said, "all models are wrong, but some are useful." Models are useful in part because the are abstract, but they should do more.

Decisionmakers want advice from economists that is useful, prudent, concise, and flexible. First and foremost, decisionmakers want useful advice. To be useful, that advice can guide strategy and/or tactics, refine the decisionmakers' thinking or understanding, and/or solve problems. Managers will often say, "I knew that would happen, but I didn't know the impact would be that big (or small)." In this sense, models can refine managers' judgment. Second, decisionmakers want accurate advice, or at least prudent advice. To enhance the advice's usefulness, it should be concise. Finally, the advice should be flexible and allow decisionmakers to calculate the cost and benefits of following their intuition if it differs from the economist's advice by incorporating uncertainty into the analysis.

The Real Estate Investment Environment

Unlike their colleagues in the stock and bond markets, institutional real estate investors have been slow to adopt the techniques of Modern Portfolio Theory (MPT). MPT embodies techniques for reducing risk in an investment portfolio by constructing portfolios along a frontier that optimizes the return per unit of risk.

This slow adoption is largely the result of the nature of the real estate market. Most institutionally owned real estate investments are traded in private markets where by definition information costs are higher than in "perfect" markets. As a result, most real estate investors operate on the assumption that the inefficiencies of a private market can be legally exploited by accessing this "inside" information and trading on it through expert intermediaries.

The traditional real estate decision process reflects a bias towards the detail and specifics of "doing the deal." It could be said that real estate portfolios are built "one deal at a time." When institutional real estate investment grew at an astounding rate during the 1980s, the focus was on acquisition and not risk management. Moreover, the acquisition process is very time-consuming. One survey found that a median time to acquire a single institutional-grade property (regardless of size) was 125 person-days.(1) The greatest attention and resources have therefore been devoted to specific transactions with less attention directed to portfolio strategy.

However, the deal-by-deal approach, even if it in fact delivered higher risk-adjusted returns (which is debatable), is not consistent with the notion of portfolio strategy. An aggregation of "best deals" could include a "great deal" on a particular property in a given market without concern for whether that deal is a good one compared to other properties in different markets.

What should distinguish investment managers over the long mn is their ability to deliver good portfolio returns repeatedly. A systematic strategy should give managers the ability to refine their judgment, dissect positive and negative outcomes, and learn to adjust their strategy from past performance. This is not characteristic of the investment professional who has the lucky coincidence of being in the right place at the right time. While strategy and tools may be transferable from one manager to another, luck is not transferable.

However, forces have been converging upon the institutional real estate investment industry that are rendering the traditional "do-the-deal" portfolio-management style ineffectual. The size of many real estate portfolios has grown to be a multiple of their annual acquisition budgets. The huge size of these real estate portfolios requires a shift away from deal-level decisionmaking to portfolio-level decisionmaking.