Reliability of HCAD land valuations

[Dave Fagundes]

Was talking to a colleague recently who mentioned the importance of land value as a benchmark when estimating the value of property. This led me to think about what the best way is to estimate land value apart from the value of any improvements. The first thing that occurred to me is the HCAD estimate, which splits its property tax base value into both land value and improvement value.

That said, I've heard that HCAD property tax valuations can differ from going market rates for a lot of reasons. This could mean that the land value listed in property tax records is not reliable. Then again, it seems it should be a lot easier to estimate the land value of a parcel than the improvement value. Land is going to be relatively uniformly valued in a given submarket, putting aside any features that make it especially attractive (ocean view) or unattractive (backing up to a bayou). Improvement value seems like it would be a lot harder to estimate and may be the major reason why HCAD valuations can differ from market rates.

So is HCAD's land valuation figure a reliable reflection of the land value of a given parcel in Harris County? If not, is there a better way to figure the value of land separately from improvements? One option would be to figure the $/sf of any recently sold vacant lots in the area, but in Houston (inner loop at least) there aren't a lot of those for sale, plus sales figures aren't public in Texas. Thanks for any thoughts on this.