significantly across properties. Sale price can be viewed as the sum of the present value
of the income that the property is producing each year. However, net rents are hard to
determine; they are often determined by complicated rent step-up procedures. Price also
depends on the credit quality of the current tenant, the terms remaining on the current
leases (for example 2 years vs. 15 years), the probability of the current tenant renewing
the lease, the cost of re-tenanting the property, etc. Such information is not available in
this data set and therefore any potential factors related to the lease structures of the
properties are not captured in the current model.
The reported results in Table 15 reveal that the estimated coefficients on the
structural attributes are of the predicted sign and statistically significant in most of the
market regressions. The coefficient on AGE is negative and significant in all but three of
the market models. The coefficient on AGE2 is positive, as expected, and significant in
10 out of the 15 markets. The coefficients on SQFT SQFT2, LANDSQFT, and
LANDSQFT2 are all positive and highly significant. The coefficient on PARKING is
positive, but significant in only three of the markets. The coefficient on FLOORS is
positive and significant in all but two of the regression models. The coefficient on
CONDITION BA has negative estimated coefficients, which are also significant in three
out of the 15 regressions. The coefficient on CONDITION AA is positive and significant
in approximately 50 percent of the models.
The estimated coefficient on the primary variable of interest representing
replacement exchange, EXREPL, is positive and significant in 14 out of the 15
regressions. Also the estimated coefficients on EXREPL tend to be much larger than with
the apartment models. The lowest estimated coefficient is in Oakland, 0.101, while the