A periodic blog dedicated to providing commentary and encouraging debate on topics in Economics and Finance.

About Me

Age: 26 Occupation: Private Equity

Friday, April 13, 2007

House Price Fundamentals: By the Numbers

I've been getting a lot questions in emails from readers about the relationship between fundamentals and prices, and what exactly it means when their real estate agent or stockbroker tells them "the fundamentals are strong" and that they should buy.

A Word on Fundamentals

In the medium and long-run, fundamentals always drive asset price valuation (the value of your stock or real estate). In real estate prices, the main fundamental drivers are incomes and interest rates. When your broker says the fundamentals are strong, she is indeed correct: interest rates are low and incomes are increasing (though not excessively so). However, there is not some line at which fundamentals become "officially strong" and put a lead foot on the infinite asset price accelerator. While this is the way its generally presented in the media or by your brokers, it's ridiculous if you sit for a moment and think about it. If interest rates decline by 1%, what happens? Will the "magic" of 5% mortgage interest cause prices to rise 5 times faster than incomes rise for all time?

The obvious answer is no. The buyer still has the same amount of money in hand to dedicate to housing. The price he is willing to pay will adjust upward in the long-run until the monthly payment approximates what it did at the higher interest rate. After that happens, even though the "fundamentals are strong," it's ridiculous to assume that prices will continue to increase indefinitely unless "Average Joe's" income rises proportionately.

If that's true, then what happened to housing? Answer: Speculation

Notice that I use the term "long-run" above. In the short run, speculative greed and fear of missing out (i.e. you'll miss out if you don't buy now... you'll be priced out forever) can cause significant distortion in markets. Speculators are the "technical" traders in the stock market or the flippers in the real estate market who live by the mantra "the trend is your friend." They don't buy based on an asset's underlying value, but instead they speculate on price swings in search of a short-term "easy money" profit.

To an extent, speculation is a self-fulfilling prophecy. As the rise in prices cause more and more speculators to pile on, with each participant searching to cash-out off of a "greater fool," the rate of price increase accelerates. In the end, the "smart money" takes its profits off of the table before the inevitable fall in prices, and they sell to the "greatest fools" who are left holding the bag as profit-taking feeds on itself, increasing the rate of price decline. Eventually, the market falls out of favor with speculators, fundamentals dominate and the market regulates for a time. The effects of speculation are well documented in history, from "Tulipomania" in 1600's Holland to the dot.com bubble of the late 1990's and the housing mania of 2004-2006, with plenty of stops in between.

What Should House Prices Be Today Given the Fundamentals?

I'll start with the good old days of April, 2000. When the boys were boys, the economy was strong and the "beer flowed like wine." Based on the U.S. Census data taken at the time, the median household income was $42,000 and the median house price in the U.S. was around $125,000. At a rate of 8.2%, a 100% mortgage payment on the "median" house was around $935. Through April 2006 (data is not yet available for April 2007, but I'll provide estimates later), incomes rose 19% according to the U.S. Bureau of Labor Statistics, meaning we can expect, based on incomes, that the median mortgage payment would also rise 19% to $1113 per month.

Given the lower interest rates in April 2006 (6.43%), that $1113 per month can buy a lot more house than it used to(which is why interest rates are a second major "fundamental.") Adjusting for both wages and interest rates, the "fundamental" value of a median U.S. house would be $177,343 in April,2006, up from $125,000 in April, 2000 - a perfectly valid 42% appreciation in house prices (before speculators stepped in and screwed the darn thing up). The manic fever that swept the country (we're always looking for the easy ticket to fortune) drove median prices all the way up to $211,578, based on changes in the OFHEO Home Price Index from our base value of $125,000, creating a 19.3% ($34,000) gap between the 2000-based fundamental value and the actual price of a house. So, when your real estate broker or David Lereah tries (or tried) to use the "fundamentals" to support current house prices, the numbers simply don't add up, and haven't for two years. (Unless you believe that some new housing technology has made houses 20% better at providing shelter... or all houses jumped 20% in size over the last 7 years - most of the houses out there today were around back then.)

Click on the picture for a larger image.

As you can see in the graph above, from 2001 to 2004, buyers were able to get a better "deal" than in 2000, and buying could have been considered a fantastic value in historic terms (especially in 2003). However, as prices began to rise, at first due to better fundamentals, speculation took hold, driving the prices way above their 2000-associated "fair value." Based on estimates for April 2007, we can already start to see house prices converge with the fundamentals.

Assuming (generously) flat house prices from this time last year, and a 3.0% increase in wages (most recent BLS estimate), and given the current drop in mortgage rates to 6.20%, we can already see the house prices begin to converge with the historic fundamentals.

Click on the picture for a larger image.

If we assume mortgage rates stay the same (given the mess in mortgages and inflationary pressures, I believe they're more likely to rise, but just for argument's sake), and incomes rise by 3%, it will be 4 more years until the median family budget buys as much house as it did back in 2000... if house prices don't fall appreciably.

Click on the picture for a larger image.


Divergences from fundamentals don't often last that long... it's more likely that prices will decline slightly to aid the correction. A 3% annual decline in house values and a 3% annual increase in incomes will erase the market distortion by April 2009. At any rate, it looks like a particularly ugly time to buy a house right now.

Click on the picture for a larger image.


For my next post... I'm going to give you the real goods on "Buy vs. Rent," and explain why you can "throw your money away" just as easily by paying too much for a house as you can by renting. In fact, it's not a matter of throwing your money away at all... we do so in both instances. It's a matter of whether you're throwing more money away buying or renting.

- eternitus

--------------------------------------------------------------------------------------------
At the request of John M., you can find the numbers for the charts below.

A note on the 2000 adjustment (for the UberNerds out there)... the Census data actually says that the median house price was $119,600 in 2000. However, since the Census doesn't have a yearly home price index, there is no way to get the Census yearly % change. So, I used OFHEO's % changes from this figure. Because of the differences in methodology between the Census and OFHEO, doing that only yielded a value of $202,000 in April 2006, below the $210 - $225 k range reported. Obviously, OFHEO and the Census don't precisely agree (statistical samples rarely do). So, I used a figure close to $119,600 ($125,000) that will get us to a reasonable result when using OFHEO's price adjustments.

At any rate, changing the starting point to $119,600 or to $130,000 would not change the % valuation gap, and only reduces / increases the estimated dollar amount of the gap by $1,000, so it's not material to the analysis... mostly cosmetic.


An interesting fact to note is that prices jumped by 69% as measured by changes in the OFHEO Home Price Index from 2000 to 2006. To put that in perspective, median house values took 40 years to double before this period.


Click on the picture for a larger image.

15 comments:

Anonymous said...

As usual, an absolutely superb analysis. I wish MSM pundits would take the time to do the math like this. We'd all be better off.

eternitus said...

Wow,
8 minutes... that's a new record for a comment on this blog.

Thanks Ronjon.

Anonymous said...

Just what the NAR doesn't want anyone to know. . . thanks eternitus, I am indebted to you

Anonymous said...

Is this a finance blog or a housing blog? Everyone knows real estate is the best investment that you can make, hands down. I wish people like you'd just go away and quit dragging my property values down.

eternitus said...

Chester,
Definitely a finance blog... but it seems that housing is the central issue right now (especially if you're in my age group).

For most of us, we're just looking for a place to live. Housing can be a good investment, but in every case, the quality of the investment is determined by the price at which you buy.

Buy high and your profits die.

Anonymous said...

What qualifications do you have to make you an expert on housing?

MyTwoCents said...

Eternitus,

Great analysis. I'm curious, is there a way to extend this approach to multiple years? I.e. can you make April of 1999 your starting point and recalculate the divergence between fundamentals and actual price?

I guess the reason I ask is, what if someone contends that April 2000 is not a valid starting point? Can you then just find the same differential from multiple starting years and then average the differentials?

I never did study statistical analysis too closely so I don't know if an approach like this would lend anything to the conversation.

Thank you again for this great blog!

My $0.02.

eternitus said...

My $0.02,
You make an excellent point. The base year makes all the difference!

Notice that if I use 2006 as my base year, 2007E looks pretty good. The reason that I chose 2000 (vs. 2003, which would make the current gap even larger) was to give "the benefit of the doubt" to housing.

I wanted to base the analysis on a good year with exceptionally strong economic conditions... This makes sure that a detractor couldn't use the argument that I was sandbagging housing by comparing a currently good economic year with an economic depression.

eternitus said...

To anonymous (really should put a name down):

My credentials (not that you need them to use published statistics to do an analysis... just need to know math, have an analytic mind, and have the gift of common sense.)

Degree: Economics and Statistics

3 years as a financial analyst

The PE fund that I work for currently invests 50% in real estate (we quit doing condos / multifamily in 2005). Have been doing offices and hotels since then.

eternitus said...

Continuing the previous comment (My $0.02)... it may be possible to refine the analysis by finding what % of a family's budget was dedicated to housing over a 10 year period, to smooth out natural year-to-year fluctuations.

But I simply don't have enough time to do that.

So, as a compromise, I chose to pick a good year that I think would provide a fair comparison in economic terms.

Anonymous said...

Nice blog, added to favorites. Perhaps the idea that needs to be conveyed is the old addage "Teach a man to fish and he eats forever". I did not have a problem understanding the math and the graphs were very well done. But if you showed your calculations, others could plug in thier own numbers and reach thier own conclusions.

I would add another addage for "Chestertheinvestor" who has more dollars than sense, "A fool and his money are soon parted".

John M

MyTwoCents said...

eternitus,

Any chance you can do a post showing how 10K invested in stocks/bonds (perhaps with leverage) can provide better long term results than realestate?

Maybe over a 15 year period?

People often assume real estate performs better than stocks bonds but I don't think they realize the inherit leverage involved.

Just a thought for a post topic.

Great blog!

My $0.02

eternitus said...

John M and My $0.02,
Thanks to the words of encouragement.

John M, I'll post my spreadsheet before I do my next post. However, the important thing to know is:

1. Mortgage Data - Mortgage x

2. House appreciation data - OFHEO Home price index - Q2.

3. Initial house price - U.S. Census - 2000 (adjusted upward slightly so the OFHEO price changes resulted in a current value similar to what we are seeing currently... may be because Census Data was actually using 1999 values).

3. Wage Data - BLS.

So with the historical stuff, I really didn't make any material assumptions.

Again, thanks for the words of encouragement. It makes the time spent on the larger posts worthwhile.

Anonymous said...

Mr. Eternitus, Quite a superb blog. You are extremely knowledgeable. Keep up the invaluable information. I look forward to your next entry.

eternitus said...

Thanks Anon... BTW, Mr. Eternitus is my dad. Feel free to call me eternitus.

My $0.02 - in my Buy vs. Rent post coming up, your issue will be addressed indirectly. (What you're really talking about is project analysis - classic finance).