Imran's personal blog

May 19, 2011

How to tell if a house is a good deal

Filed under: Uncategorized — ipeerbhai @ 2:02 pm

Previously, I had posted about expected value, but got feedback that the post was too academic.  Here’s my attempt to simplify the information.

Background

House prices are set by the market — the price of a house is whatever a seller is willing to sell it for and a buyer willing to buy it for.  A house that was built for $100,000 in materials may sell for $1,000,000 or for $10,000 — it has nothing to do with the actual costs of the house.

Currently, the US is in the middle of the “Great Recession” — and since house prices depend on the overall economy to a large degree, house prices are depressed.  how depressed?  They haven’t been this low, adjusted for inflation, since 1982 — at least near Seattle — it may be different where you are.

The question

These low prices are making people think about buying a house as an investment now — and selling it when the economy recovers.  They want to know — is this house a good deal?  I’ll define a “good deal” as a house you can buy now, and profit from by selling when the economy recovers.  So, you need to find out what the price would be for the house when the recovery happens.  How does one estimate this?

Methods of finding the answer

Well, there’s a few different ways to estimate this.  Most economists use a model based on rents — and there’s evidence that way works pretty well in some situations.  Some economists use a model based on regressions of previous prices — and there’s evidence that model works in some situations.  A third way is to try and figure out when the bubble started, and assume the price will return to pre-bubble values.

The easiest way

All three ways work — in some cases.  The simplest is to use the “what was the price before the bubble” method — just lookup the house’s assessed value from before the bubble, and assume that’s what the price will be when the economy recovers.  In Seattle, the bubble began in early 2004 — so an assessed value from 2003 should work for this method in Seattle.  County Tax offices usually have this data — just call them up, give them an address, and ask for the 2003 tax valuation on the property.  Viola — you have one way of knowing.

The safest way

The other two methods require some math skills.  The rents method — find a similar house for rent, and assume the rent payment is the same as a mortgage payment, and do a reverse amortization, then add 20%.  This is the “Peter Schiff” method — well, a really simplified version.  It’s a good model — Schiff was one of the few economists who correctly predicted the bubble, and is Ron Paul’s economic advisor — Goldman Sachs also uses a method similar to this.  If you don’t know how to do a reverse amortization( and most people don’t), you can find an online mortgage calculator, and it can tell you.  This model is safe, because in the worst case, you could rent out the house for your payments — others already have.

The most accurate way

The final method is the regression.  This method is wrongly blamed for the great recession — banks switched to using it, but using it incorrectly, causing them to over-value houses.  Of course, they did it wrong — purposely! A regression is a little hard to set-up, but gives great results provided you feed it good data. Most regressions use 20-30 years of data, and are pretty accurate. Unless you’re cheating. Banks, being incentivized by the Republican Gramm-Leach-Bliley act, did the regressions using only 2-3 years worth of data — against all common sense. They argued that the new tranches and CDO’s that they designed were safe, “AAA” rated, and thus any data prior to the introduction of CDOs was no longer relevant. So, they threw away all the old data — the same data that told them houses were overvalued, and new loans should not be made — and did regressions with only data from when CDO’s were designed. It was this throwing away of data that caused the great recession, worsened by the perverse incentives of Gram-Leach-Bliley act.

Anyway, a regression probably provides the best estimate.  But it’s also the hardest to do — you need to get comparable house prices for 20-30 years, normalize them, then run a regression using specialized statistical software.  Frankly, unless you’re willing to spend a few weeks learning R, eViews, SQL, and other things like that, this method is very hard to use.  You should go down to your local university and find a smart economics student, and ask them to do it for you.  This is the method many real-estate hedge funds use, and done with correctly, this method explains the most and can give you the biggest profits.

Final Answer

So, the final answer for estimating the non-recession price of a house is to use all three methods, and pick the lowest number. Any price below that is probably a good deal. Be warned — the regression method shows house prices can still fall before returning to normal. House prices adjusted for inflation were this bad or worse surprisingly often in the 60s, 70s, and 80s. No method can predict how low prices will go. No method will work if the US economy collapses. If the political administration makes a horrible mistake, all methods will be wrong. But, that’s not important. Eventually, the economy will recover — perhaps not in our generation, but eventually. When it does, those who got good deals now will be sitting pretty. The regression method gives us one piece of data — when things turn, they turn fast. If you get a good deal, don’t worry about the little losses you may take now — as long as you can meet your payments, you’ll make a killing about 2 years after the economy recovers.

FYI — The regression models predict that house prices in Seattle will bottom this year ( 2012 ), and that the economy will begin recovery in 2014.   House prices should be fully recovered to their normal values around 2016, provided the 2014 recovery happens as expected.( House prices are dependent on the unemployment rate and how long its ben high.  I view recovery when unemployment materially decreases.  )

Advertisements

Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: