Market Analysis SES 0549500 Lecture 8 Rena 14-15 October 9 &11 Office Markets Presented by: Raymond G. Torto
Exercise 2 Review: Effect of Price Increase in Asset Market Asset Market: Valuation Rent $ Space Market: Rent Determination D 0 11% OAR D 1 8% OAR R* SR R** P 1 Price $ LR P** P* Q* Q* * Stock (SF) Asset Market: Construction C* Space Market: C** Stock Adjustment Construction (SF) CBRE Page 2
How Many Office Using Jobs?
Macro Demand: 3 Approaches Aggregate Approach Defining office employment Disaggregate Approach Industry-occupation matrices Econometrics CBRE Page 4
Aggregate Approach CBRE Page 5
The North American Industry Classification System (NAICS) Office Employment Definitions 11 Agriculture, Forestry, Fishing, and Hunting 21 Mining 22 Utilities 23 Construction 31-33 Manufacturing 42 Wholesale Trade 44-45 Retail Trade 48-49 Transportation and Warehousing 51 Information 52 Finance and Insurance 53 Real Estate and Rental and Leasing 54 Professional, Scientific and Technical Services 55 Management of Companies and Enterprises 56 Administrative and Support and Waste Management and Remediation Services 61 Educational Services 62 Health Care and Social Assistance 71 Arts, Entertainment and Recreation 72 Accommodation and Food Services 81 Other Services (except Public Administration) 92 Public Administration CBRE Page 6
Office Space Usage: Employment by NAICS Office Employment* in Dallas and Chicago, 1989 Dallas Chicago Standard Industrial Classification (SIC) Total (thousands) Office (thousands) Total (thousands) Manufacturing 184.7 16.2 499.1 49.4 Mining 17.4 10.3 1.3 0.6 Construction 47.5 0.6 93.8 0.4 Office (thousands) Transportation, Communication, and Utilities (TCU) 92.4 7.1 148.5 6.2 Trade 287.9 28.1 613.6 51.1 Finance, Insruance, and Real Estate (FIRE) 122.9 122.9 246.0 246.0 Services 314.8 105.8** 730.2 227.0 Total Private 1067.6 291.0 2332.5 580.7 * Those employees occupying separate office space from on-site manufacturing *adapted from DiPasquale and Wheaton (1996) ** includes advertising, computer and data processing, credit reporting, mailing and reproduction, legal and social services, membership organizations, engineering and management services. CBRE Page 7
Disaggregate Approach CBRE Page 8
Econometric Approach
Review? Regression Analysis Explained Univariate or Multivariate Linear R2 Coefficients and statistical significance Absorption and vacancy over time CBRE Page 10
U.S. Office Fundamentals Completions and Absorption, msf Vacancy, % 150 Forecast 21 100 18 50 15 0 12-50 9-100 6-150 3 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: CBRE EA, Office Outlook, Q2 2012. CBRE Page 11 Completions (L) Net Absorption (L) Vacancy (R)
Stock Flow Model CBRE Page 12
Models are estimated empirically using real estate data together with Economic data Option #1: reduced form forecast just evaluate and forecast rents with a model that has either a trend or Economic Demand variables. Option #2: forecast rents and construction together. Assume market clearing. Base forecast conditional on Economic variables. Option #3: add in vacancy and assume that markets clear slowly = more variables and more equations. Better forecasts! CBRE Page 13
Model #1: Unconditional Univariate R = 3.23 +.92R -1 -. 09T (2.5) (22.1) (.6) R 2 =.933 No trend (T) in real office rents (.09 is not significant). Rents depend an awful lot on last periods rents! R* = (3.23-.09T)/ (1-.92) = $39! (long term steady state rents in real dollars) CBRE Page 14
Model #2: Conditional Univariate R = 13.2 +.94R -1 -.06 FIRE +.023 SER (4.1) (36.2) (3.6) (2.9) R 2 =.942 Trend is replaced with office employment. Does that work? Rents still depend an awful lot on last periods rents! Why is it that growth in FIRE jobs creates no rent growth? Why does Service grow have positive impact? Who forecasts FIRE and SER? CBRE Page 15
Model #3: Conditional Rent/Construction Multivariate The estimated demand/rent equation becomes: R = 7.6 +.94R -1 +.04FIRE +.02SER -.00013S -1 (2.9) (17.1) (2.1) (3.2) (-3.9) R 2 =.976 For supply: C = 449 + 39.7R -10 -.007S -1 (.9) (2.9) (-2.2) R 2 =.41 S = S -1 + C System will forecast rents and construction and the stock of space given FIRE and SER forecasts. Who forecasts FIRE and SER? That s what is meant by conditional. CBRE Page 16
Forecasting Lessons When supply adjusts quickly to prices or rents, then little is to be gained from a model that jointly forecasts the two Just use a Univariate model The slower supply responds and the more gradual prices and rents adjust, then the more you need to forecast both sides of the market to capture its momentum and cyclic swings. CBRE Page 17
Specifying an Office Model CBRE Page 18
Estimating a Lease Rent Index: Hedonic Index (average annual$/sqft over lease term) log(r) = α 0 + α 1 SQFT + α 2 GROSS1 + α 3 GROSS2 + α 4 TERM + α 5 HIGH 1991 n + α 6 NEW1 + α 7 NEW2 + Σ β i D i + Σ δ j S j (1) i=1979 j=1 Var iable Denver Cincinnati Houston San Fr ancisco Washington Constant 1.8153 2.0887 2.0700 2.4211 2.2169 Square Feet 1.08e-06 1 3.35e-07 1-8.42e-07-4.57e-06-1.03e-07 1 G1 0.0952 0.0993 0.0574 0.0172 1 0.1420 G2 0.0728 0.0315 1 0.0316 1 0.0633 0.1177 1 Term 0.0290 0.0196 0.0203 0.0260 0.0120 High 0.1048 0.1293 0.0586 0.1119 0.0361 Dummy 1979-0.0681 1 na 0.0082 1 na na Dummy 1980 0.2860 na 0.1290 0.0790 1 na Dummy 1981 0.4775 na 0.3480 0.3664 0.0684 1 Dummy 1982 0.5992 0.0468 1 0.3925 0.4847 0.1872 Dummy 1983 0.5468 0.1305 0.3300 0.4193 0.2176 Dummy 1984 0.5394 0.1385 0.1995 0.4879 0.3996 Dummy 1985 0.5402 0.1128 0.1646 0.4525 0.4113 Dummy 1986 0.3556 0.1378 0.1314 0.3408 0.4422 In economics, hedonic regression is a revealed preference method of estimating demand or value. It decomposes the item being researched into its constituent characteristics, and obtains estimates of the contributory value of each characteristic. CBRE Page 19
Some Conclusions Even Today CBRE Page 20
Variance: Distribution of Forecast Outcomes A forecast is the mean value of the variable(s) being forecast. Any forecast has a probability distribution surrounding it. The further out you go the wider is the forecast probability distribution of possible outcomes. Why? CBRE Page 21
What is Risk: Historic Variability vs. Forecast Uncertainty Historic Variability NOI Forecast Uncertainty TIME CBRE Page 22
What Determines Confidence Band Width? A market with wide historic swings will tend to generate wider confidence bands in the future unless you can explain these swings accurately. A poor model (low fit) means you do not understand the forces affecting the market. What you don t know = risk. Low quality data, missing observations, a short historic data series, no variables that capture what really drives the market = a poor model. CBRE Page 23
Ask for the back test! Or the fitted vs. actual CBRE Page 24
Forecasting Submarkets CBRE Page 25
Forecasting Submarkets: Its Complicated! Submarkets represent close if not perfect substitutes. Modeling choice among close substitutes is almost impossible with traditional demand models: large number of cross price elasticity's, aggregate consistency Economic data may be endogenous space drives job creation rather than the reverse. There is no economic data available for submarkets that offers consistent annual time series. Hence data on exogenous factors is not available. Submarket performance data (rents, vacancy, absorption) is always very noisy. Small erratic projects and decisions have big impacts! CBRE Page 26
THIS IS WHY WE GO TO THE MICRO ANALYSIS OF OFFICE MARKET OR ANY PROPERTY MARKET RENA CHAPTER 16 MICRO ANALYSIS RENA CHAPTER 17: CASE STUDIES CBRE Page 27
Office Markets in the Major Global Financial Centers
Office Employment in: NY, London, HK Office-Using* Employment (in thousands) 2000 1800 1600 Forecast 1400 1200 1000 800 Hong Kong NYC London 600 400 200 0 1988.1 1990.1 1992.1 1994.1 1996.1 1998.1 2000.1 2002.1 2004.1 2006.1 2008.1 2010.1 2012.1 2014.1 2016.1 2018.1 2020.1 2022.1 Sources: CBRE EA, Moody's, Oxford Economics. *Includes both business & financial services. CBRE Page 29
Just What is Office Employment? Growth in business services (IT, Media, law ) has far outpaced that in financial services and now dominates tenancy - in all 3 markets: A structural change? Financial and business services employment (x 1000): 1997 versus 2011 1,400 1,200 1,000 800 600 400 1997.4 2011.4 200 0 Fin Serv Biz Serv Fin Serv Biz Serv Fin Serv Biz Serv Hong Kong New York City London Sources: CBRE EA, Moody's, Oxford Economics. CBRE Page 30
Office Space Demand Growth in: NY, London, HK Office Net Absorption (4Q MA) as a % of Office Stock 4% 3% Forecast 2% 1% 0% Hong Kong NYC London -1% -2% -3% 1988.1 1990.1 1992.1 1994.1 1996.1 1998.1 2000.1 2002.1 2004.1 2006.1 2008.1 2010.1 2012.1 2014.1 2016.1 2018.1 2020.1 2022.1 Source: CBRE. CBRE Page 31
New Office Supply: Larger Markets (NY) Intrinsically Less Volatile than Smaller(HK) 4% Office New Supply/Construction as a % of Office Stock 3% 2% Forecast 1% 0% Hong Kong NYC London -1% -2% 1988.1 1990.1 1992.1 1994.1 1996.1 1998.1 2000.1 2002.1 2004.1 2006.1 2008.1 2010.1 2012.1 2014.1 2016.1 2018.1 2020.1 2022.1 Source: CBRE. CBRE Page 32
Office Vacancy Rates in NY, London, HK 20 Office Vacancy Rate (%) 18 16 Forecast 14 12 10 8 6 4 Hong Kong NYC London- Total 2 0 1988.1 1990.1 1992.1 1994.1 1996.1 1998.1 2000.1 2002.1 2004.1 2006.1 2008.1 2010.1 2012.1 2014.1 2016.1 2018.1 2020.1 2022.1 Sources: CBRE EA, Moody's, Oxford Economics. CBRE Page 33
The Source of Vacancy Volatility Varies In NYC most vacancy movement is due almost exclusively to demand fluctuations not supply. In London there is moderate volatility in demand and supply, but they are totally uncoordinated creating considerable volatility in vacancy. Hong Kong s building booms generate potential vacancy volatility but demand is remarkably coordinated. Hence vacancy volatility actually is less than NY, London: σ vac = 2.6 versus 4.4 (NY), 3.3 (London) Diversification? R 2 (NY, London)=.69, R 2 (HK, NY/London)= -.29 CBRE Page 34
Office Rents in: NY, London, HK 1.5 Nominal Office Rents (Indexed to Q4 2000) 1.0 Forecast Q4 2000 = 0 0.5 0.0 Hong Kong NYC London-Total -0.5-1.0 1988.1 1990.1 1992.1 1994.1 1996.1 1998.1 2000.1 2002.1 2004.1 2006.1 2008.1 2010.1 2012.1 2014.1 2016.1 2018.1 2020.1 2022.1 Source: CBRE. Note: HK & NYC use prime rents. NYC uses the TW Office Rent Index. CBRE Page 35
Office Rents: Historical Patterns Historically, rents in each market are identically and negatively correlated with vacancy: NY (-.69), London (-.68), HK (-.61). But HK rents react more strongly (a larger coefficient) to its smaller vacancy swings. This generates much more rent volatility. Historic trends in all three markets remarkably similar: 2.5% to 2.9% per annum nominal growth. Little diversification value to NY and London (R 2 =.88) but some between each and HK (R 2 =.37). CBRE Page 36
The Next Decade All three markets should see below average vacancy and renewed rent growth in the next 3 years baring another global financial crisis! Longer term, rent levels in NY and HK reaches new high s in comparison to the previous peak (2007). London s growth stalls after 2017 and rent levels never exceed the previous (2007) peak. Longer term rent growth in HK also slows from historic trend as demand growth slows (relocations to mainland cities) but supply does not drop sufficiently. As HK matures its volatility starts to approach that of the other markets. CBRE Page 37
U.S. Office Fundamentals Completions and Absorption, msf Vacancy, % 150 Forecast 21 100 18 50 15 0 12-50 9-100 6-150 3 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: CBRE EA, Office Outlook, Q2 2012. CBRE Page 38 Completions (L) Net Absorption (L) Vacancy (R)
Square feet/worker: Changes in professional Occupation Ratio: Rental Cost of Occupancy, Technology? Occupied Square Feet Per Worker 225 220 215 210 205 200 195 190 185 180 175 TW Rent Index, 2004$ psqft 40 38 36 34 32 30 28 26 24 22 20 1980 1981 1983 1985 1987 1988 1990 1992 1994 1995 1997 1999 2001 2002 2004 2006 2008 2009 Occupied sqft Per Worker TW Rent Index, 2004$ CBRE Page 39
Rental Elasticity of Office Space Demand CBRE Page 40
Inventories are Best Predictor of Industrial RE Demand Where Do Global Firms Decide to Store Goods? 150 Industrial Net Absorption (millions sf) Change in Bus. Inventories (billions 2005 $) 40 100 50 30 20 10 0 0-50 -100 Correlation=0.769-10 -20-30 -150-40 1990q1 1991q1 1992q1 1993q1 1994q1 1995q1 1996q1 1997q1 1998q1 1999q1 2000q1 2001q1 2002q1 2003q1 2004q1 2005q1 2006q1 2007q1 2008q1 2009q1 Industrial Net Absorption Change in Real Business Inventories CBRE Page 41
Rooms Sold vs. Real GDP: GDP and Room Rates Are All That Matter! National Hotel Market 3000000 2500000 2000000 1500000 1000000 500000 0 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Rooms Sold Real GDP CBRE Page 42