CNBC Recap & Analysis: Tax Assessment, Mortgage Refis, Valuations, Conforming Loan Limits

August 12, 2011

by Scott Sambucci


We tackled several topics in my five-minute segment CNBC’s Street Signs on Wednesday  – “Housing: A Silver Lining” – so here’s some deeper analysis of the on-air arguments and assumptions.

Valuation methodology is the acute problem for most of today’s housing market problems – refinancing, demand stimulus, foreclosures.  The valuation and appraisal process is broken, creating the problems raised by Brian Sullivan and Mandy Drury about tax assessment and refinancing obstacles.

1. Tax assessments & property valuations

Dependence on lagging transaction data is the kiss of death in any housing valuation model. (See: Fungal Houses and Tails of Price Change)

Brian argued that municipalities are over-valuing homes for tax purposes – much higher than what the banks are willing to value for purposes of refinancing. We have record low interest rates and refinancing applications are higher, but underwater homeowners can’t refinance and they’re in turn stuck with inaccurately high tax bills.

Here’s a case example from my neighborhood in Davis, CA: 571 Jerome Street

$575,000 (Sold Price, March 2008)
$549,000 (Listed for Sale, June 2011)
$542,000 (Current List Price, August 2011)
$530,000 (Current Median Ask Price for all homes in this Price Quartile)
$447,000  (Current Zestimate; Range: $334-534k)

2010 Tax Assessment:     $585,000

The last assessment was completed for 2010, using transaction data from 2009 when prices where in the $600,000 range:

Ask & Sold Prices since 2009: Second Price Quartile for Davis, CA, zip code 95616 (Price Quartiles are 25% price bands in local markets, providing a more precise analysis of "like" homes to a subject property)

The $585,000 valuation is clearly too high because:

  • There’s been volatility and a downward shift of home prices in this community
  • Market trends changes in this price quartile
  • The property is currently for sale at $542,000 and hasn’t sold

Keep in my that you can’t have it both ways – a lower valuation for tax purposes and higher valuation for selling purposes – and be prepared to accept the seasonal upswings too. Notice the short-run lift in home prices since the early Spring this year.  On a real-time basis, valuations have risen back to the $550-560,000 range with an inflection point coming as the spring and summer selling seasons unwind that will push the valuation back down in the second half of 2011.

Most importantly, the housing market’s real-time volatility is not captured using most tax assessor valuation methods.

(Brian – send me an email with your city and zip code.  I’ll shoot you over a market report you can use with your municipality to argue for a reduction in your tax assessment.)

2. Refinancing current, but underwater, mortgage borrowers

At the start of the segment, CNBC played a sound bit about $2 trillion of performing mortgages paying 1% or higher than the current mortgage rate available.  Mandy asked about running credit checks and income verification to validate refinancing opportunities for current, but underwater borrowers.  Sounds simple right? This is the great tragedy of the HAMP program.

Here’s the fundamental problem – If you do a refi with a house that’s underwater, where does the missing loan principle go?

If you’re current on your mortgage, the bank has no incentive to provide a modification, refinance, or write-down. You’re asking the banks take a lower interest rate on a smaller principle for a current borrower. Would you do this deal if you were a lender?  How would shareholders respond to that on an earnings call? (“You did what!?”…)

A newly-released FHFA Working Paper – “The HAMP NPV Model: Development and Early Performance” explains the HAMP methodology succinctly in this context:

Testing modifications for positive NPV generally eliminates borrowers who are very unlikely to be foreclosed upon or who have substantial positive equity, because in both cases the mortgagee or lien-holder is unlikely to suffer meaningful losses in the no-mod case. The NPV test also eliminates borrowers for whom a modification does not meaningfully reduce their prospects of foreclosure.

This exemplifies how the system is set up to encourage borrowers to strategically default so they can seek modification, if the borrower is willing to take a hit on their credit rating.  Brent White at the University of Arizona covers this topic extensively.  (See: “Underwater and Not Walking Away” and “Take this House and Shove It: The Emotional Drivers of Strategic Default“)

3. Back to the valuation problem

A transaction is an agreement on price and a disagreement on value. – Trader’s Proverb

The crisis yields an opportunity for the housing market to reset it’s valuation methodology based on real-time indicators – active market indicators, rent-to-own ratios, mortgage rate sensitivity analysis, demographic shifts – all of which are largely ignored by the lenders and servicers in their instructions to appraisers and BPO providers.

Housing is illiquid compared to other asset classes. The assets are idiosyncratic and trade infrequently, and when they do trade, buyer and seller biases and preferences exist – motivation, financing availability, price floors and ceilings, and property requirements.

Appraisers are instructed to solve this problem – establish market value without a willing buyer and seller of the subject asset.  The asset itself isn’t trading and the appraiser is required by outdated regulations to use a specified number of closed and active market comparables, both of which leave their analysis incomplete. Even attempts such as the 1004MC form introduced by Fannie Mae in 2009 include minimum and subjective analysis related to real-time market conditions.

Every appraisers would like access to more data to provide better analysis because they know they’re only as good as their last appraisal.

4. Effects of the upcoming FHA Conforming Loan Limit change

Brett Kornfield and David Bugajski at Discern Analytics (using housing data provided in part by Altos Research) researched this question in great detail back in June. In short, the effects will be minimal.  Additionally, these effects will diminish further as the active inventory from June sells off throughout the summer, leaving even fewer homes available for sale affected. Here’s an excerpt from their June 10 Applied Strategy report – “The Diminimis Impact of Conforming Loan Limit Changes”:

Source: Discern Analytics (

My final thoughts

  • Fix the valuation problem with more real-time analytics.
  • Good luck solving the moral hazard problem and incentives for strategic walkaways.
  • No, I don’t wear ties to work – we’re in Silicon Valley.   Heck, it was lucky that I found a sport coat to wear.
  • Yes, I know.  Mom always said I had a face for radio.


Previous post:

Next post: