Yale School of Management Introduction to Real Estate History and Concepts.

Yale School of Management

Introduction to Real Estate

History and Concepts

Yale School of Management

The Dynamics of Real Estate Markets

Real Estate Finance Spring 2005

Yale School of Management

From Pro Forma to Stochastic Processes

• Pro Forma risk analysis

• Cash flows depend upon:– Scenarios– Probability assessments

• Discount rates depend upon– Systematic vs. unsystematic risk drivers

• Is there any way to incorporate all of this?

Yale School of Management

Simulation Tools

• Value drivers:– Rents– Vacancies– Expenses

• Drivers of value drivers:– – –

Yale School of Management

Simulation Methods

• Requires structure/model– Rent processes– Vacancy processes– Interest rates– Covariance estimates

• Model: Sivitanides, Torto, Wheaton (2003)– MSA/aggregate structural relationships

Yale School of Management

Forward Looking?

• Rational Model: Efficient markets– Agents anticipate future conditions and trends

• Myopic model– Agents react to immediate conditions– Rational explanation?– Muth model

Yale School of Management

Are Cycles Rational?

Time

Index Values (USD)

0.8

20

1

2

3

4

5

6

7

8

910

11.210.8

3.1

Dec1977

Jun2004

Dec1978

Dec1979

Dec1980

Dec1981

Dec1982

Dec1983

Dec1984

Dec1985

Dec1986

Dec1987

Dec1988

Dec1989

Dec1990

Dec1991

Dec1992

Dec1993

Dec1994

Dec1995

Dec1996

Dec1997

Dec1998

Dec1999

Dec2000

Dec2001

Dec2002

U.S. Inflation NCREIF Property TR U.S. LT Gvt TR

Yale School of Management

Cobweb Model 1

Yale School of Management

Cobweb Model 2

Yale School of Management

STW Analysis

• Interest rates matter

• Spreads and Cap Rates not forward-looking

• No trend towards efficiency

Yale School of Management

Rents and Vacancies

Yale School of Management

Patterns:

• Autocorrelation

• Inter-dependence

• Mean reversion

Yale School of Management

Cap Rates and Interest Rates

• C = NOI/P e.g. Before Tax Yield• Why Negative?• Why Positive?

Yale School of Management

Hypotheses

• Inflation hedge.• Lower future growth.• GDP changes.• Recent trends: dropping since 2000

– 9.5 to 8.5 RCA– 8.5 – 7.5 NREI– NCREIF no change

• What about diversification?• Trends in the equity market?• Sentiment?

Yale School of Management

Survey of Institutional Investors+/ - Assets Allocated to Real Estate over next 2-3 Years

0

10

20

30

40

50

60

70

80

I ncrease Decrease Stay the Same Uncertain Prefer not toanswer

Q5

Fre

qu

en

cy

Yale School of Management

STW Analysis

• Cap Rates moved by interest rates

• Historical analysis remains reliable

• Other factors that could explain structure?– Strategy?– Other investments?

Yale School of Management

Back to Simulation

• CF depends on:– Rents, vacancies

• Prices depend upon – interest rates– Growth expectations– inflation

Yale School of Management

Simple Simulation:

• Rents follow a random walk• R(t) = R(t-1) + e(t)• E(t) is a random error• Spreadsheet simulation straightforward• Take last qtr rent, add a normal error term to

it, then move forward one cell for ten cells.• Do this 100 different times and look at

range of outcomes.

Yale School of Management

Problems

• Random walk assumption

• Normal errors and positive rents

• Don’t know std of error

• Don’t know if random walk makes sense

• What to do?

Yale School of Management

More Complex Recipe

• If rents autocorrelated– Estimate an autoregression:– R(t) = a + bR(t-1) + e(t) SAVE ERRORS– Take R(0) as today’s rents– R(1) = a+bR(0) + e* where * means random

draw from saved errors.– Move to the next cell

Yale School of Management

Ultimate Recipe

• Include other variables in estimation stage

• Vector auto-regression – Allows rents to depend on past vacancies– Allows vacancies to depend on past rents– Allows them to depend on past interest rates

• Also allows simulation of extreme cases

Yale School of Management

VAR Forecasts