Long-run equilibrium for the Greater Paris Office Market ; Rental and Demand adjustments

1
Long-run equilibrium for the Greater Paris
Office Market Rental and Demand adjustments

• European Real Estate Society Annual Conference
• Bucharest, 2014

Catherine Bruneau and Souad Cherfouh
Professor, University Paris I Pantheon Sorbonne
and Paris School of Economics PhD student,
University Paris I Pantheon Sorbonne and BNP
Paribas Real Estate
2
INTRODUCTION
3
Motivation for the research

• Despite its leading position in Europe, limited
  research on the Greater Paris rental office
  market
• Analysis of the Parisian market dynamics (focus
  on the rental and demand equations)
• Extend the existing office market research to the
  French case to the really recent past
• Combine cointegration techniques to a multiple
  structural break approach

4
MODELING STRATEGY
5
Literature and methodology

• Theoretical background ? multi-equation framework
  
• Rosen, 1984 Wheaton, 1987
• Wheaton et al. 1997 Malle 2010 etc
• Objective provide a full comprehension of the
  main underpinning market dynamics
• 3 behavioural equations
• Demand(economic activity and rents)
• Supply(economic activity, vacancy rate,
  construction costs, interest rates)
• Rents(vacancy rates)

6
Literature and methodology

• Econometric methodology ? cointegration approach
• Hendershott et al. 2002 Brounen and
  Jennen, 2009
• McCartney, 2012 Hendershott et al., 2013
  etc
• Reduced-form equations
• Justification most economic, financial and real
  estate series are non stationary
• Objective provide a relevant framework able to
  account for specific characteristics of space
  markets ? well-known slow adjustment

7
Literature and methodology

• Econometric methodology ? multiple structural
  break approach
• Perron, 1989 Banerjee and Urga,1995
  Gregory and Watt, 1996 Gregory and Hansen,
  1996
• Justification no cointegration found between the
  variables of interest
• Objective account for the structural changes
  that may affect the long run equilibria through
  shifts in the mean
• Nonlinear approach in the space market literature
  
• Englund et al. 2008 Brounen and Jennen,
  2009 Hendershott et al., 2010 etc

8
Rent modeling

•  

(1)
(2)
9
Demand modeling

• Long-run model
• Short-run model

(3)
(4)
 
10
THE GREATER PARIS OFFICE MARKET
11
The Greater Paris office market

• Largest office market in Europe in terms of stock
  (53 m sm) and in terms of turnovers (over 2 m sm
  take-up/annum since 2000)
• Second largest market in terms of investment
  after London ( 9.6 b/annum since 2000 vs
  11.8 for London)
• Major market with specific characteristics
  compared to the London market
• Higher level of office space centralization in
  France
• Economic growth prospects strongly reliant on a
  relatively stable private and public demand
• Shorter terms of rental agreements

12
Vacancy and rental movements
Figure 2 Real rent and vacancy rate in the
Greater Paris office market (1991-2012)
13
Movements in office demand fundamentals

• Figure 3 Occupied stock and real rents in the
  Greater Paris office market (1991-2012)

• Figure 2 Occupied stock and office employment in
  the Greater Paris office market (1991-2012)

14
EMPIRICAL RESULTS
15
Estimation procedure

• ? Three-step procedure
• 1. Structural break identification ? pragmatic
  approach
• No closed-form solutions for the limiting
  distribution of the unit root test statistics in
  the case of multiple breaks
• Graphical examination of the long-run residuals ?
  choice of the dates that best corresponds to
  changes in the mean of the residual.
• Critical values obtained from Monte Carlo
  simulations.
• 2. FMOLS estimation of the cointegrating
  relationships for both sub-systems with their
  respective breaks.
• 3. OLS estimations of the ECM within each
  sub-system

16
Rent modelling

• Coefficient for vacancy rate significant and
  expected sign
• All breaks are justified by referring to recent
  developments of the parisian market the related
  coefficients have the expected signs.
• For example
• 1994Q4 ? sharp rent correction consecutive to the
  bubble burst in the early 1990s
• 2000Q4 ? demand crisis 2 vacancy rate for
   relatively low  rental values ? correction in
  the distorsion

• Table 1 Long-run equilibrium rent model

Variable Coefficient Coefficient Coefficient t-Statistic
         
Vacancy rate -0.08 -16.3 -16.3
-0.26 -10.9 -10.9
-0.15 -4.9 -4.9
0.27 12.5 12.5
-0.13 -6.5 -6.5
0.06 3.0 3.0
Intercept 5.07 109.9 109.9
         
         
N 88 Adjusted R² Adjusted R² 0.93
         
Notes Dependent variable ln(Real Rent). ,
, indicate significant at 10, 5 and 1
respectively.
17
Rent modelling

• Error correction term is significant long run
  causality from the vacancy rate towards the rent
  (Granger, 1988)
• Speed of rental adjustement 29
• Relatively high speed of adjustement compared to
  other European markets
• Shorter lease lenght
• Higher market turnover
• Autoregressive structure of the rents

• Table 2 Short-run adjustement model

Variable Coefficient Coefficient Coefficient t-Statistic
         
?Vacancy rate(-2) -0.04 -6.0 -6.0
?ln(Rent(-3)) -0.32 3.4 3.4
Error correction term -0.29 -4.2 -4.2
Intercept -0.00   -0.9 -0.9
         
         
N 84 Adjusted R² Adjusted R² 0.55
Durbin-Watson 2.16      
         
Notes Dependent variable ?ln(Real Rent). ,
, indicate significant at 10, 5 and 1
respectively.
18
Demand modelling

• Coefficients for employment and rents
  significant and expected signs
• All breaks can be interpreted and the related
  coefficients have the expected signs
• For example
• 1993Q3 ? downwards correction in the supply
  crisis context
• 2001Q3 ? downwards demand correction relative to
  changing economic conditions
• 2007Q4 ? financial crisis impact

• Table 3 Long-run equilibrium demand model

Variable Coefficient Coefficient Coefficient t-Statistic
         
ln(Office employment) 1.10 23.1 23.1
ln(Real rents) -0.12 -11.2 -11.2
-0.05 -7.3 -7.3
0.02 4.3 4.3
-0.02 -4.7 -4.7
0.01 3.5 3.5
-0.02 -3.6 -3.6
0.02 5.0 5.0
Intercept 12.87 70.0 70.0
         
         
N 87 Adjusted R² Adjusted R² 0.99
         
Notes Dependent variable ln(Occupied Space).
, , indicate significant at 10, 5 and 1
respectively.
19
Demand modelling

• Table 5 Absorption adjustement model

• Error correction term is significant
• Slow speed of space adjustement 9
• Space absorption is sticky due to rigidities
• Long-term term leases
• Costs associated with lease termination

Variable Coefficient Coefficient t-Statistic
       
?Occupied Space(-1) 0.31 3.3
?Occupied Space(-2) 0.28 2.7
Error correction term -0.09 -1.8
Intercept 0.00 2.7
       
       
N 85 Adjusted R² 0.23
Durbin-Watson 1.93    
       
Notes Dependent variable ?ln(Occupied Space).
, , indicate significant at 10, 5 and 1
respectively.
20
International comparison
Table 4 Elasticity estimates for selected
European office markets

• Notes Cautious is required when comparing
  between studies with different specifications,
  sample periods and proxies.

21
Conclusion and future work

• Results summary
• Justified structural breaks allow us to capture
  cointegrating relationships
• Error correction approach confirmed as relevant
  to model space market
• Settles the traditional drivers of rents and
  demand in the long-run
• Characterizes their role in the short-run ?
  outlines the rigidities specific to the market
  structure
• On going work
• Submarket analysis ? dynamics and interactions
  between Parisian submarkets (Stevenson (2007), Ke
  and White (2014))

22
References

• Adam, Z. and Füss, R. 2012. Distangling the short
  and long-run effects of occupied stock in the
  rental adjustment process. Journal of Real Estate
  Finance and Economics 44 570-590.
• Banerjee, A. and Urga, J. 1995. Modelling
  structural breaks, long memory and stock market
  volatility an overview. Journal of Econometrics
  119 1-34.
• Brounen, D. and Jennen, M. 2009. Local office
  rent dynamics. Journal of Real Estate Finance and
  Economics 39 385-402.
• Füss, R., Stein, M. and Zietz, J. 2012. A
  Regime-Switching Approach to Modeling Rental
  Prices of U.K. Real Estate Sectors. Real Estate
  Economics 40 317-350.
• Gregory, A.W., Nason, J.M. and Watt, D.G. 1996.
  Testing for structural breaks in cointegrated
  relationships. Journal of Econometrics
  71 321-42.
• Gregory, A.W. and Hansen, B.E. 1996.
  Residual-based tests for cointegration in models
  with regime shifts. Journal of Econometrics 70
  99-126.
• Hendershott, P. H., Jennen, M., MacGregor, B.
  D. (2013). Modeling space market dynamics an
  illustration using panel data for US retail. The
  Journal of Real Estate Finance and
  Economics, 47(4), 659-687.

23
References

• Malle R. 2010. Un modèle à équations simultanées
  du cycle des bureaux en région parisienne.
  Economie et Prévision 3 93-108.
• McCartney, J. 2012. Short and long-run rent
  adjustment in the Dublin office market. Journal
  of Property Research 29 201-226.
• Perron, P. 1989. The Great Crash, the oil price
  shock and the unit root hypothesis. Econometrica
  57 1361-401.
• Rosen, K. 1984. Toward a Model of the Office
  Building Sector. Real Estate Economics 12
  261269.
• Wheaton, W.C., R.G. Torto and P. Evans. 1997. The
  Cyclic Behavior of the London Office Market. The
  Journal of Real Estate Finance and Economics 15
  77-92.
• Wheaton W. 1987. The Cycle Behavior of the
  National Office Market. Journal of the American
  Real Estate and Urban Economics Association 15
  281-299.