Planning to Retire – A Financial Autobiography – Step 1 – How Much Is that Housey In the Window?

I love owning a home. And even better, owning a home free and clear by the time we retire will make our retirement less expensive and more secure. But I personally don't see a home as an investment. It's not diversified and if we depend on the ability to cash in on it then we have to give up our freedom to choose where we live. So for now our goal is to own a home free and clear by the time we retire but not to rely on the equity in the home for living expenses during retirement.

Reducing Uncertainty

The great thing about owning our home free and clear by the time we retire is that it takes a lot of risk out of retirement. In fact, it's extremely profitable way to live. Instead of having to generate income (which is taxed) to pay rent, we get to, in effect, pay ourselves rent and we don't have to generate any income to do it! It's like getting a discount on our living expenses equal to our tax rate. It also means that as housing expenses change due to inflation, the economy, whatever, we are protected. That's a very comforting thought.

Still, owning a house isn't a completely free ride. Maintenance is still an issue as are property taxes. As anyone living in California during the 1970s can tell you it is quite possible for housing prices to explode and property taxes to go with them. This is what eventually contributed to the introduction of Proposition 13, which made it a lot safer to own a house but did a real number on California's finances [CAP]. Many states, afraid of the repercussions of Prop 13, have therefore put in place protections for the elderly against property tax increases.

So while there is no such thing as a free lunch owning our own home by the time we retire seems a pretty good idea.

Store of Value?

But the real question is – should we treat our home as an investment? Should we count on the value it will build up and expect to tap that value for income during our retirement? Treating a house like an investment seems to me to be the equivalent of buying a single stock and saying we're invested. Strictly speaking it's true but we've exposed ourselves to a lot of risk by not diversifying properly. Lots of things can happen to a particular house. For example, here is a sample of things that have happened in our upper middle class neighborhood in the last few years:

  • What apparently was a meth lab was run from a house down the block

  • Convicted sex offenders have moved into the area

  • The city is trying to re-develop a near by military base into a traffic choking sports complex with flood lights that will be on until late in the night

Any one of these things could have done a number on our individual home's value.

Besides, how exactly should we estimate the likely value of our house three decades from now? And then how are we to estimate the value of our house during retirement, not to mention estimating things like interest rates which have a strong affect on how much equity we have in the house and how easily we can extract that equity? We don't know what state we'll be living in three decades from now much less what house.

It's also fair to wonder if housing prices are necessarily predestined to go up over extended periods of time. [BBP], for example, argues that housing prices move in tandem with the portion of the population that is employed. But we know that in the future a smaller and smaller portion of the American population will be working (see the last column of Table V.A2. in [SDP]). Which is why [BBP] argues that housing prices, in the long term, are going to fall in America. Now it's only fair to point out that [BBP] makes a whole boat load of assumptions which may not turn out to be true and it ignores a lot of factors to simplify the analysis but I think the argument is interesting enough, when combined with the diversification and location issues, to make me hesitant to treat our house as an investment.

Cashing In

Once upon a time the only way we could realize the equity in our house would be to leave it, either by renting it out or selling it. These days there is a third option, a reverse mortgage. See [FRM] and [HCR] for good, but brief, descriptions of the key features of reverse mortgages.

If we plan on extracting the equity from our house by selling or renting it we have to leave the house (or have tenants, yuck). That actually seems a reasonable enough plan to me as the house we want to own now is designed for a family and when we retire it will just be the two of us. But watching my own parents make these decisions I've seen how hard it is to leave a home you've made your own over a period of decades. My parents aren't moving from their family sized home and I hope to have the same option to stay where we are for Marina and I.

If we use a reverse mortgage we can stay right where we are but at the risk of never being able to move. For example, let's say we have taken out a reverse mortgage. All the money we are lent earns interest which is due back to the lender. But the interest is secured against the equity of the house. The good news is that if the amount we owe exceeds the value of the house the lender looses, a reverse mortgage only allows the lender to get the equity in the house and that's it. But interest is due on all money lent out under a reverse mortgage and that interest is charged against the equity in the house. The end result is that a reverse mortgage can very quickly eat up all the equity in the house. So if we decide we want to move, for whatever reason, we can easily find ourselves with no equity left in the house and so no funds with which to move. In a very real sense we end up prisoners in our own home.

If we end up needing money we will be very glad that we have the option to rent or sell or to get a reverse mortgage. All of these options have their place and the world is, I believe, better off with these choices available. But these choices come at a real cost and so we hope to try and avoid that cost by planning our retirement so that we don't have to depend on the equity in our home.

Appendix – House Price Appreciation in the Seattle Area

I wanted to get a better understanding of housing prices in the Seattle area so I did some simple analysis using the following data sets (full details on these data sets is available below):

  • East Side Average – This data set measures the average prices for houses in the East Side of King County which includes Bellevue/Redmond/Issaquah.

  • King County Average – Average sales price of houses in King County, Washington.

  • King County Median – This measures the median price for houses sold in King County.

  • Seattle/Bellevue/Everett – This measures the average prices of conforming loans (e.g. Fannie Mae/Freddie Mac loans) in northern King County (including Seattle and the East Side) and South Snohomish county.

The only period of time that all three data sets overlap is 1996-2000 and two of the data sets overlap for 1996-2004, I have included below a summary of their behavior:

East Side Average

King County Average

King County Median

Seattle/Bellevue/Everett Average for Conforming Loans


10.82% (8.74%)

7.98% (3.44%)

5.12% (3.12%)

3.80% (2.59%)




4.86% (3.01%)

3.88% (2.27%)

Table 1: Geometric Average Yearly Return (Standard Deviation) in Inflation Adjusted Dollars

The difference in growth rates between the East Side and King County as a whole illustrate the old real estate rule – location, location, location. A house in the right place increases in value a lot faster than a house in the 'wrong' one. But the comparison of the King County Average and the King County Median illustrates another rule, the rich get richer faster than the poor do.

The comparison of the King County Median and the Seattle/Bellevue/Everett average is interesting because the King County Median price from 1996-2004 was always just under the conforming loan limit. So one can think of the King County Median as showing the most expensive conforming loans while the Seattle/Bellevue/Everett average shows the entire range of conforming loans. But, again, although the geographical areas aren't quite identical, I suspect it's still reasonable to conclude that the rich get richer faster.

It was also interesting to note that all of the data series contained periods of time with negative real returns. In other words, the old saw that housing only goes up just isn't true. In Seattle housing regularly loses value, it's just that over longer periods of time they tend to go up.

I have included a spreadsheet which contains all the data, analysis and a graph of inflation adjusted housing prices from the various data sets. Note, however, that this spreadsheet is in Open Office 2.0 format. To view it please download Open Office 2.0, it's open source and so available for free. Please note that this spreadsheet does include a macro and so if you get a dialog asking if you want to enable macros when you load the spreadsheet you will need to say yes in order to see the analysis results.

Adjusting For Inflation

In the original version of this article I adjusted by inflation by taking the CPI-U and removing the shelter component. The theory being that by including the shelter component I was double counting housing inflation since I was also measuring house prices. But I later realized that this didn't make any sense. The point in adjusting the housing prices for inflation is to get a feeling for what the value of a house was given the value of a dollar at a specific point in time. To measure the subjective value of a dollar (and inflation measurements are nothing if not subjective) I was best off using the full inflation index since that best measures the general 'worth' of a dollar.

What would be mildly interesting is to do the calculations again but only use the shelter component (specifically rent substitutions) as a way of stripping out of the data series the inflation component of housing. I would then be left with 'constant' "house dollars". By "house dollars" I mean dollar values that reflect how much "house" I can buy. But these dollars would be more than a little deceiving because they would only apply to buying a house. If I used the "house" dollars for anything else then their 'actual' value would be different. Yes, I know, confusing.

In any case I decided in the end to use the CPI-W index because, to be blunt, it includes the 'better off' 32% of city dwellers and thus is closer to the inflation rate Marina and I experience. I also choose non seasonally adjusted data mostly because I use annual averages not monthly averages so the seasonal adjustment (which adjusts monthly results for seasonal effects) wasn't useful. Besides only the national inflation average is calculated using seasonally adjusted data and I wanted to use the Seattle-Tacoma-Bremerton CPI-W data series which is not seasonally adjusted. One slightly odd thing is that the monthly data series (CWURA423SA0) does not provide coverage for all months but that's o.k. because it gives the annual data which is what I actually wanted. But even odder the monthly series is missing all data for years 1987-1996. But the semi-annual data series (CWUSA423SA0) does contain the annual data for that period. Just to add to the fun the monthly series goes back to 1914 but the semi-annual series only goes back to 1984. But when I compared similar years the annual data was identical. Go figure.

Sources of Data I Used

King County Annual Growth Report (KCAGR) 2001 – The KCAGR, until 2001, provided average prices of houses in the East Side and King County from the Central Puget Sound Real Estate Research Report [CPS]. The listing is available in section IV, sub-section "Sub County Areas" of [KGR]. KCAGR reports after 2001 compromised their data sets by pulling data from multiple different sources and then gluing them together which means I can't rely on their consistency of methodology, which is why I haven't used KCAGR data subsequent to that contained in the 2001 report (which itself only listed information through 2000).

Office of Federal Housing Enterprise Oversight's (OFHEO) Housing Price Index [HPI] – The OFHEO is a government agency responsible, amongst other things, for determining if Fannie Mae and Freddie Mac are sufficiently capitalized. To help it do that the OFHEO generates the Housing Price Index (HPI) which is an index that measures the cost of housing in various areas. The problem with the HPI is that it is calculated based exclusively on conforming loans, that is, loans issued and/or serviced by Fannie Mae and Freddie Mac. In 2006, for example, the maximum principal amount of a single family conforming loan is $417,000 [FMC]. The HPI is broken up by various metropolitan areas, the one I used covers Seattle-Bellevue-Everett. The index data is generated on a quarterly basis so to calculate the yearly price changes I averaged the results over the four quarters and then calculated (average year 2 – average year 1)/(average year 1) where year 2 is the year I want to know the growth rate in house prices for and year 1 is the previous year.

Washington Center for Real Estate Research (WCRER) [WCR]- WCRER is run by Washington State University and collects and publishes housing data about counties in Washington State including median prices for King County.

Alternate Data Sources I Didn't Use

U.S. Department of Housing and Urban Development & U.S. Department of Commerce American Housing Survey [AHS] – This is the survey equivalent of a colonoscopy, no secret remains hidden. They do have a report covering both Seattle and Everett (ala the HPI) but unfortunately they organize their data in a less than useful way. They break house prices down into price segments, the final one of which is "all housing costing over $300,000" which is too low to be interested in areas like Seattle. They do include median prices but the WCRER data set is better for me since it focuses exclusively on King County. The WCRER data set also covers a slightly longer period of time, running from 1995 where the AHS dataset only seems to go back to 1997.

U.S. Department of Labor, Bureau of Labor Statics (BLS), Consumer Price Index (CPI) – [CRR] explains the gory details but the upshot is that when the consumer price index, the usual measure of inflation, is calculated a "shelter" component is included in the calculation to measure changes in housing costs. Once upon a time this index was calculated by sampling both rents and actual house prices. But in the early 80s the BLS changed the methodology to measure rent and "rent equivalent". A "rent equivalent" is how much the owner of a home would get if they rented the home out. From an economics point of view this approach makes a ton of sense because the true 'value' of a home should be more or less equal to the income stream it can generate, e.g. rents. But when you live in odd times, odd rules apply. For example, in the East Side one can trivially rent a house that would sell for $1.2 million for only $2,500/month. So, in so far as the CPI is concerned, that house is worth $2,500 in rent equivalent even though buying the house would cost many times that per month assuming 20% down and a 30 year fixed rate mortgage, not to mention property taxes and maintenance. So the bottom line is that the CPI is not a good measure of housing price behavior because it isn't actually measuring house prices. An interesting consequence of this is that the CPI will completely miss the inflationary aspects of a housing bubble which I suspect means that the current CPI is a lousy measure of inflation.


[AHS] American Housing Survey (AHS), U.S. Census Bureau & The Department of Housing and Urban Development.

[BBP] Martin, Robert F. "The Baby Boom: Predictability in House Prices and Interest Rates", Division of International Finance, Board of Governors of the Federal Reserve System, 11/2005.

[CAP] Moore, Stephen, "Proposition 13 Then, Now and Forever", Cato Institute, 7/30/1998.

[CPS] "Central Puget Sound Real Estate Research Report", The Central Puget Sound Real Estate Research Committee.

[CRR] "Consumer Price Indexes for Rent and Rental Equivalence", U.S. Department of labor, Bureau of Labor Statics, 3/29/2002.

[FMC] "Fannie Mae Announces 2006 Conforming Loan Limit of $417,000", Fannie Mae, 11/29/2005.

[FRM] "Facts for Consumers – Reverse Mortgages: Get the Facts Before Cashing In On Your Home's Equity", Federal Trade Commission, 6/2005.

[HCR] "Top Ten Things to Know if You're Interested in a Reverse Mortgage", U.S. Department of Housing and Urban Development, 6/28/2004.

[HPI] Housing Price Index (HPI), Office of Federal Housing Enterprise Oversight.

[KGR] "2001 King County Annual Growth Report", King County Washington, 2001.

[SDP] "V. Assumptions and Methods Underlying Actuarial Estimates", 2002 OASDI Trustees Report, Social Security Administration, 3/26/2002.

[WCR] Washington Center for Real Estate Research, Washington State University.

One thought on “Planning to Retire – A Financial Autobiography – Step 1 – How Much Is that Housey In the Window?”

  1. Enjoyed your series on your retirement plans. If you haven’t had a chance you might want to take a look at “Retire on the House”. I found it to be full of a lot of good ideas about leveraging home equity in retirement. I’m not associated with the book (other than this brief review.

    You might also want to check out the notion of “reality retirement planning” which takes into account fact that retirees spend less as they age. Not mainstream thinking by any means, but I think the idea has merit and is overlooked by many.

    I first came across it in this article by Ty Bernicke and have even put up a reality retirement calculator on my site.


Leave a Reply

Your email address will not be published. Required fields are marked *