Comments on “Land Prices and Fundamentals”

Comments on “Land Prices and Fundamentals”

by Koji Nakamura and Yumi Saita

Masahiko Shibamoto

Research Institute for Economics and Business Administration Kobe University

December 1, 2007

This paper ...

• exmamines long-run relationship between macro economic fundamentals and the weighted-average land price indicators, which are supposed to be appropriate indicators in the context of the macro economic analysis.

• cointegration between the land price and DPV.

⇒ explosive ‘bubble’ doesn not exist, although the actual land prices can be deviate the theoretical land prices.

• error-correction analysis:

convergence of actual land prices to the long-term equilibrium level

My comments:

1. Fundamentals

‘myopic expectations of the future income’

⇒ (defensible specification?) Economic fundamentals fluctuate dramatically!!

2. Empirical Approach

3. Are weighted-average land price indicators really appropriate in the context of the macro economic analysis?

Comment 1: Fundamentals

• Discouted Presented Value (DPV) of Land Price

P_t = Y_t + E_tP_t+1

1 + r_t , and r_t = i_t + τ_t + RP_t

⇒ P_t = Y_t

i_t − g_t^e + τ_t + RP_t (6)

• Authors use the estimation specifications in line with the theory of land price determination. However, I think that the theoretical values obtained by authors has some problems, although authors point out in p. 40.

• g_t^e: constant growth rate (static expectation for future income growth)

growth rate of the quarterly nominal GDP filtered by HP filter (λ = 100 instead of λ = 1600)

⇐ Is people’s expectation really myopic?

Can you use HP filter as the proxie? Are there some persuasive reasons and/or previous studies?

smoothing parameter λ = 100 ?

• cf. Campbell and Siller (1987, JPE) estimate DPV based on the VAR model.

• Risk Premium: RP_t = RP = 6% ⇐ Constant !!

bubble: low

before and after bubble: high

g

_t^e

-20 -10 0 10 20 30

60 65 70 75 80 85 90 95 00

GROWTH_ANNUAL HPTREND_100 HPTREND_1600

g

_t^e

㪄㪉 㪄㪈 㪇 㪈 㪉 㪊 㪋 㪌 㪍

㪈㪐㪎㪐㪇㪋 㪈㪐㪎㪐㪈㪇 㪈㪐㪏㪇㪇㪋 㪈㪐㪏㪇㪈㪇 㪈㪐㪏㪈㪇㪋 㪈㪐㪏㪈㪈㪇 㪈㪐㪏㪉㪇㪋 㪈㪐㪏㪉㪈㪇 㪈㪐㪏㪊㪇㪋 㪈㪐㪏㪊㪈㪇 㪈㪐㪏㪋㪇㪋 㪈㪐㪏㪋㪈㪇 㪈㪐㪏㪌㪇㪋 㪈㪐㪏㪌㪈㪇 㪈㪐㪏㪍㪇㪋 㪈㪐㪏㪍㪈㪇 㪈㪐㪏㪎㪇㪋 㪈㪐㪏㪎㪈㪇 㪈㪐㪏㪏㪇㪋 㪈㪐㪏㪏㪈㪇 㪈㪐㪏㪐㪇㪋 㪈㪐㪏㪐㪈㪇 㪈㪐㪐㪇㪇㪋 㪈㪐㪐㪇㪈㪇 㪈㪐㪐㪈㪇㪋 㪈㪐㪐㪈㪈㪇 㪈㪐㪐㪉㪇㪋 㪈㪐㪐㪉㪈㪇 㪈㪐㪐㪊㪇㪋 㪈㪐㪐㪊㪈㪇 㪈㪐㪐㪋㪇㪋 㪈㪐㪐㪋㪈㪇 㪈㪐㪐㪌㪇㪋 㪈㪐㪐㪌㪈㪇 㪈㪐㪐㪍㪇㪋 㪈㪐㪐㪍㪈㪇 㪈㪐㪐㪎㪇㪋 㪈㪐㪐㪎㪈㪇 㪈㪐㪐㪏㪇㪋 㪈㪐㪐㪏㪈㪇 㪈㪐㪐㪐㪇㪋 㪈㪐㪐㪐㪈㪇 㪉㪇㪇㪇㪇㪋 㪉㪇㪇㪇㪈㪇

㪟㪧㪶㫋㫉㪼㫅㪻㪈㪇㪇 㪟㫇㪶㫋㫉㪼㫅㪻㪈㪍㪇㪇 㪈㩷㫐㪼㪸㫉㩷㪸㪿㪼㪸㪻 㪊㩷㫐㪼㪸㫉㫊㩷㪸㪿㪼㪸㪻 㪌㩷㫐㪼㪸㫉㫊㩷㪸㪿㪼㪸㪻

Comment 2: Empirical Approach

• Specification of the Cointegration Regression 3,4 (p.26)

p_t = β₀ + β₁T rend_t + β₂N P V_t + e_t (15)

Economic interpretation that you relax the restriction on the coefficient of the DPV of land of one ?

β₂ = 1 imply Equation (6).

β₂ > 1, β₂ < 1 ???

• Should test whether or not the coefficient of the DPV is one (instead of 0).

(In fact, its coefficient is significantly defferent from one in many case.)

• Can the trend term represent the structural decline in demand for lands due to the changes in the economic structure?

deterministic? I think that this is represented by the stochastic.

• Estimation of Cointegrating Vectors

p.30 “DOLS is the ordinary least squares method with leads and lags of the dependent variables” or

日本語バージョン

p.29

「説明変数の階差のラグ項を 加えた」

???

⇒ the OLS method with leads and lags of the first difference of the regressors!!

• Estimation of Error-Correction Models

Δp_t = β₀ + β₁EC_t−1 + β₂ΔN P V_t + β₃Δpop_t + β₄Δc_t + _t

I recomend IV estimation, instead of OLS (because ΔN P V_t etc is not exogenous), or OLS estimation including some lags of Δpop_t, ΔN P V_t and Δc_t instead of ΔN P V_t,Δpop_t, and Δc_t.

Comment 3: Are the weighted-average land price really appropriate?

(Figure 1) Long-Term Trends of Land Price Indicators⁵

0 50 100 150 200 250 300 350

1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 SNA Land Value

Weighted-Average Land Price Index Published Land Prices(Official Publication) Urban Land Price Index

㧔

1980=100

㧕

year Sources: Cabinet Office, “National Accounts”; Ministry of Land, Infrastructure and Transportation,

“Published Land Prices,”; Japan Real Estate Institute, “Urban Land Price Index.”

• “If we used the official land price indicators, we would underestimate the impact of the land price fluctuations for high land price areas · · · ”

⇐ I think that, if we used the weighted-average land price indicators, we would underestimate the impact of the land price fluctuations for low land price areas, and overestimate the impact of the land price fluctuations for high land price areas. I think that this is arbitrary.

Is it a problem so much that the weights of the official land price indicators are in line with the population weights?