Returns, Volatility, and Information Transmission Dynamics in Public and Private Real Estate Markets

Size: px

Start display at page:

Download "Returns, Volatility, and Information Transmission Dynamics in Public and Private Real Estate Markets"

Moris Perry
5 years ago
Views:

1 Returns, Volatility, and Information Transmission Dynamics in Public and Private Real Estate Markets by David Ling and Andy Naranjo University of Florida For presentation at: NCREIF s Summer Conference This document was presented during the NCREIF Summer Conference.

2 Background & Motivation Returns in both private & public commercial real estate (CRE) markets should be driven at least in the long-run by CFs produced by underlying properties But private & public CRE often produce significantly different returns (and risk) At least according to standard indices This document was presented during the NCREIF Summer Conference.

3 The GREAT DEBATE: Are equity REITs stocks or real estate? This document was presented during the NCREIF Summer Conference.

4 This Ongoing Debate is Misguided Publicly traded REITs: are not private direct CRE investments are not general equity securities (as proxied for by, say, the S&P5) are real estate securities that buy and sell properties in the private market To quote Michael Hudgins: REITs are, ultimately, just another way to access the longer-term return characteristics of the underlying private real estate This document was presented during the NCREIF Summer Conference.

5 Why Differences in Risk-Return Performance? Numerous factors may cause REIT prices to move differently than MV of underlying properties especially in short-run Leverage (amplifies returns) s in market value of liquidity and/or transparency s in general stock market & RE sentiment (non-rational expectations) This document was presented during the NCREIF Summer Conference.

6 So What Do We Do? Focus on two questions:. Do REIT managers or managers in the private market do a better job of selecting/timing markets, selecting properties, and/or managing properties?. What is dynamic short-run (- months) relation between private & public CRE markets? Ex., do public markets predict returns in private markets? This document was presented during the NCREIF Summer Conference.

7 Benchmark Private Market Returns Primary source of return data is NCREIF Transaction Based Index (TBI) Capital gain component of TBI in each quarter is based only on constituent properties in NCREIF database that sold that quarter NCREIF TBI indices available at national level back to : for core property types: multifamily, office, industrial, and retail Dictates beginning of analysis period This document was presented during the NCREIF Summer Conference.

8 REIT Data Initial sample obtained from CRSP-Ziman database ( Merged with Compustat to obtain quarterly accounting data necessary to unlever REIT returns at the firm level Total assets, total debt, interest expenses, preferred dividends, etc. See paper for details Availability of Compustat This document was presented during the NCREIF Summer Conference.

9 Measuring Relative Performance Comparing public & private CRE returns directly requires several adjustments:. Remove effects of leverage from firm-level REIT returns Unlevering is done quarterly (see paper for details). Exclude REITs that do not invest in core property types. Use same property-type weights as found in NCREIF index to create our aggregate core REIT index, and. Adjust NCREIF TBI returns downward to reflect management expenses TBI returns are reported gross of asset management fees; REIT returns are reported net of all firm-level management expenses We assume the difference in advisor/management fees to be 8 bps per year Difference between gross & net (of management) total returns on NCREIF s ODCE fund is 6 bps per year This document was presented during the NCREIF Summer Conference.

10 Figure : Cumulative Total Returns- Levered vs. Unlevered REITs.55 % levered Levered returns Difference = 58 basis points 8.7 % unlevered.86% levered Unlevered Returns Unlevered Equity REIT Returns Panel C: Mean Return Standard Deviation Auto- Correlation Aggregate: All-Properties 8.7%.%. Apartment.8%.5%. Office.7%.58%. Industrial.%.65%.5 Retail.%.%.6 This document was presented during the NCREIF Summer Conference.

11 Next Step: Removing Non-Core REITs property type indices created using CRSP-Ziman property type designations multifamily, office, industrial, retail Aggregate Core Properties REIT Index created from these property type indices using same weights as NCREIF index This document was presented during the NCREIF Summer Conference.

12 Figure 5: # of Core REITs as % of Total 7.% 6.% 5.%.% Indicates importance of excluding noncore REITs from analysis.%.%.%.% % Core Focused REITs This document was presented during the NCREIF Summer Conference.

13 Cumulative Return % Finally! Public vs. Private Returns (Figure 6) 5.% 5.%.% 5.%.% 5.%.% 5.%.% 5.%.% Figure 6: Cumulative Unlevered Returns on Core Properties: REITs vs. NCREIF TBI Unlevered Core Properties REIT Index:.% annualized return Difference = basis points annually Core Properties TBI Index: 8.8% annualized return Unlevered Equity REIT-Core Unlevered NCREIF-TBI Core Prop This document was presented during the NCREIF Summer Conference.

14 Cumulative Return % Cumulative Return % Cumulative Return % Cumulative Return % 5.% 5.%.% 5.%.% 5.%.% 5.%.% 5.%.% Results by Property Type: Public vs. Private Market Returns Figure 7: Apartment Returns Difference: - bps 5.% 5.%.% 5.%.% 5.%.% 5.%.% 5.%.% Figure 8: Industrial Returns Available soon on NAREIT website Difference: + bps Unlevered Apt REITs Unlevered NCREIF-TBI Apt Unlevered Industrial REITs Unlevered NCREIF- TBI Industrial Figure : Office Returns Figure : Retail Returns 5.% 5.%.% 5.% Difference: + bps 5.% 5.%.% 5.% Difference: + bps.%.% 5.% 5.%.%.% 5.% 5.%.% 5.%.%.% 5.%.% This document Unlevered Retail REITs Unlevered NCREIF-TBI Retail Unlevered Office was REITs presented during Unlevered the NCREIF- TBI Office NCREIF Summer Conference.

15 Results Sensitive to Sample Period Lack of TBI indices prior to requires us to exclude late 8s/early s Annualized Total Returns CRSP-Ziman equity REITs.7%.66%.78% NCREIF NPI -.6%.6% 7.% Difference.7%.%.% Excluding - from analysis decreases measured outperformance of equity REITs Adding would decrease cumulative REIT outperformance Equity REITs:.%; NPI;.% In? Equity REITs: 5%; NPI:.% This document was presented during the NCREIF Summer Conference.

16 Additional Adjustments that Would Improve Return Comparisons? Omit REITs from core that are too (?) focused on development Development is a noncore activity; may distort comparison with NCREIF managers Adjust for differences in liquidity Would enhance relative performance of REITs Adjust for differences in (MSA) location of properties This document was presented during the NCREIF Summer Conference.

17 Stage : Dynamic Conditional Analysis- Do Public Mkts Predict Private Mkts? Employ vector autoregressive models (VARs) to examine dynamic relation among private & public market RE returns. REIT returns = f(lagged REIT returns, lagged TBI returns, and macroeconomic control variables). TBI returns = f(lagged TBI returns, lagged REIT returns, and macroeconomic control variables) This document was presented during the NCREIF Summer Conference.

18 Wait! Don t We Know Public Markets Predict Private Markets? Yes when fundamental variables are excluded from analysis But ability of REITs to predict TBI returns may be due to omitted fundamental variables correlated with TBI returns i.e., REIT returns are not predicting private market returns; rather, fundamentals are predicting both REIT & private market returns But changing fundamentals are more quickly incorporated into liquid REIT prices This document was presented during the NCREIF Summer Conference.

19 So, an Important uestion is. Do REIT returns contain RE specific information useful in predicting private market returns? This document was presented during the NCREIF Summer Conference.

20 Baseline Dynamic VAR Results: Without Asset Pricing Control Variables REIT returns predict private market returns in subsequent quarters This document was presented during the NCREIF Summer Conference.

21 What if We Add Standard Control Variables? uarterly asset pricing control variables: yield on -month U.S. Treasury securities (TBILLM) slope of Treasury term structure of interest rates (TERMSPREAD) spread between yields on BAA rated and AAA rated corporate bonds (DEFSPREAD) rate of general inflation (INFL) excess return on market portfolio (MKT) remaining Fama-French risk factors (SMB and HML), augmented by a return momentum factor, (MOM) Pastor and Stambaugh s () liquidity innovation factor (LI_PS) This document was presented during the NCREIF Summer Conference.

22 Highlights of Dynamic Regression Results After adding controls for fundamental economic variables lagged REIT returns are not predictive of TBI returns Exception: apartments Suggests no additional predictive power in REIT returns beyond standard asset pricing control variables Explanation? REITs are not embedding real estate specific information This document was presented during the NCREIF Summer Conference.

23 Some Preliminary Results on MSA Location (with Andy Naranjo & Ben Scheick) Assume (for sake of discussion) that REIT managers outperformed -. Is this performance difference attributable to better: market (MSA) selection? individual property selection within (MSA) markets? property/asset management? or some combination of the three? This document was presented during the NCREIF Summer Conference.

24 What Do We Do? Collect the following data from SNL on an annual basis for each property held by an equity REIT during 6-: property owner (institution name) property type geographic location (MSA) acquisition date sold date book value initial cost, and historic cost Could do quarterly back to This document was presented during the NCREIF Summer Conference.

25 What Do We Do? Sort each property held by a REIT--and its book value--into a MSA Book value: historical cost of property & improvements minus accumulated depreciation Do this separately for all core property types Example: Apartments Compute % of book value of apartments owned by apartment REITs in each MSA at beginning of each year This document was presented during the NCREIF Summer Conference.

26 Percent of Portfolio REIT vs. NCREIF Apartment Allocations 6.%.%.%.% 8.% 6.%.%.%.% Washington D.C. - Apartments Private Public This document was presented during the NCREIF Summer Conference.

27 Percent of Portfolio REIT vs. NCREIF Apartment Allocations 6.% Houston - Apartments 5.%.%.%.%.%.% Private Public This document was presented during the NCREIF Summer Conference.

28 Percent of Portfolio REIT vs. NCREIF Apartment Allocations: 8.% 7.% 6.% 5.%.%.%.%.%.% Chicago - Apartments Private Public This document was presented during the NCREIF Summer Conference.

29 What (Else) Do We Do With Geography? Also calculate MSA concentrations of REIT-owned apartments based on () # of properties & () adjusted cost of each property SNL adjusted cost: max of book value, initial cost, and historic cost Will also do for gateway cities This document was presented during the NCREIF Summer Conference.

30 So How Does This Reweighting Affect NPI Returns: 6-? Based on book value Difference from NCREIF APT NPI minus reweighted NPI Difference from NCREIF IND Difference from NCREIF OFF Difference from NCREIF RET No effect + bps annually + bps annually - bps annually Mean.% -.% -.%.% Median -.% -.% -.%.65% Std Dev.8%.8%.6%.7% Min -.8% -.5% -.7% -.88% Max.7%.5%.6%.66% This document was presented during the NCREIF Summer Conference.

31 -.% -.8% -.6% -.% -.%.%.%.%.6%.8%.%.% Difference from NCREIF APT Differences by uarter Can be Large! NPI APT vs. Reweighted NPI APT.% annual in one qtr. -.% -.% -.8% -.6% -.% -.%.%.%.%.6%.8% Difference from NCREIF IND NPI IND vs. Reweighted NPI IND -.% -.5% -.% -.5%.%.5%.%.5%.% Difference from NCREIF OFF NPI OFFICE vs. Reweighted NPI OFFICE NPI RETAIL vs. Reweighted NPI RETAIL -.% -.5%.%.5%.%.5%.% Difference from NCREIF RET.% annual in one qtr. 6.%+ annual in one qtr. This document was presented during the NCREIF Summer Conference.

32 So How have we fine tuned the horse race Is this performance difference attributable to better: market (MSA) selection? individual property selection within (MSA) markets? property/asset management? This document was presented during the NCREIF Summer Conference.

33 What (Else) Will We Do? Run pooled, time-series, regressions at the firm level using REIT data to determine: the extent to which MSA concentrations increase our ability to explain firm-level REIT returns which MSA exposures increased/decreased returns and how these MSA effects have varied over time This document was presented during the NCREIF Summer Conference.

34 Returns, Volatility, and Information Transmission Dynamics in Public and Private Real Estate Markets by David Ling and Andy Naranjo University of Florida For presentation at: NCREIF s Summer Conference This document was presented during the NCREIF Summer Conference.

Public versus Private Commercial Real Estate Returns:

Public versus Private Commercial Real Estate Returns: MSA Allocations, Property Selection, and Performance Attribution By: David Ling, Andy Naranjo, and Ben Scheick Investors can own commercial RE through