In One Up on Wall Street, Peter Lynch recommends housing as the best first investment:

Because of leverage, if you buy a $100,000 house for 20 percent down and the value of the house increases by five percent a year, you are making a 25 percent return on your down payment, and the interest on the loan is tax-deductible. Do that well in the stock market and eventually you’d be worth more than Boone Pickens.

As a bonus you get a federal tax deduction on the local real estate tax on the house, plus the house is a perfect hedge against inflation and a great place to hide out during a recession, not to mention the roof over your head. Then at the end, if you decide to cash in your house, you can roll the proceeds into a fancier house to avoid paying taxes on your profit.

The customary progression in houses is as follows: You buy a small house (a starter house), then a medium-sized house, then a larger house that eventually you don’t need. After the children have moved away, then you sell the big house and revert to a smaller house, making a sizable profit in the transition. This windfall isn’t taxed, because the government in its compassion “gives you a once-in-a-lifetime house windfall exemption. That never happens in stocks, which are taxed as frequently and as heavily as possible.

Because of the deductions, this works for individuals, not professional investors, of course.

How does this strategy perform compared to 401k, index funds, and other low-risk passive instruments available to individuals? What are cumulative returns over 20-30 years?


Then at the end, if you decide to cash in your house, you can roll the proceeds into a fancier house to avoid paying taxes on your profit.

The problem is that the book was written in 1989. That comment is no longer true; that part of the tax law changed in the 1990's.

Also in 1989 the maximum amount that person could put in an IRA was $2,000 and hadn't been raised for almost a decade and wouldn't be raised for another decade. Roth accounts didn't exist; nor did HSA's or 529's. Most people didn't have a 401K.

You are asking to compare what options we have today compared to what was available in the late 1980's.

For me except, for the years 2001-2005 and 2010-2015, the period from 1988 until now has had flat real estate values. Still the current values haven't returned to the peak in 2005. The score is 11 great years, 17 flat or negative. I know many people who during the 1990's had a zero return on their real estate.

  • Could you clarify which tax law you think changed? There's no capital gains on a principal or second residence, and you can roll profits from one investment property to the next as far as I know. (If that's wrong, it would be great to know!) – user32479 Nov 24 '15 at 5:03
  • @Brick -- In the U.S., there is a large exemption on capital gains on principal residences. But single people in moderately priced areas, and married people in expensive areas often face capital gains taxes on home sales. – Jasper Dec 26 '15 at 8:26

Thinking of personal residence as investment is how we got the bubble and crash in housing prices, and the Great Recession. There is no guarantee that a house will appreciate, or even retain value. It's also an extremely illiquid item; selling it, especially if you're seeking a profit, can take a year or more. '

Housing is not guaranteed to appreciate constantly, or at all. Tastes change and renovations rarely pay for themselves. Things wear out and have costs. Neighborhoods change in popularity. Without rental income and the ability to write off some of the costs as business expense, it isn't clear the tax advantage closes that gap, especislly as the advantage is limited to the taxes upon your mortgage interest (by deducting that from AGI).

If this is the flavor of speculation you want to engage in, fine, but I've seen people screw themselves over this way and wind up forced to sell a house for a loss.

By all means hope your home will be profitable, count it as part of your net wealth... but generally Lynch is wrong here, or at best oversimplified. A house can be an investment (or perhaps more accurately a business), or your home, but -- unless you're renting out the other half of a duplex,which splits the difference -- trying to treat it as both is dangerous accounting.

  • Housing appreciates constantly. Right now it's above the 2007 level (eg research.stlouisfed.org/fred2/series/MSPNHSUS ). With the tax benefits, it may outstrip the alternatives. That's why I'm asking effective ROI. – Anton Tarasenko Nov 22 '15 at 17:12
  • Which investment has risk-adjusted returns greater than those of housing? – Anton Tarasenko Nov 22 '15 at 18:44
  • 1
    @AntonTarasenko: Housing where? The thing is that an index fund is the same no matter where you are when you buy it, but buying (or selling) a house in San Francisco is very different from buying a house in Duluth. – BrenBarn Nov 22 '15 at 18:56
  • 1
    @AntonTarasenko: "Housing appreciates constantly." - no, it doesn't. It may appreciate over the long run, but if you buy at the peak in a bubble and the bubble bursts, it can take years for your house to regain the value for which you bought it. Bad if you need the money. Even worse if you went for the high leverage Lynch recommends, find your mortgage underwater and have to foreclose. Leverage on volatile assets can be dangerous. For some reason, every generation seems to need to relearn that. – Stephan Kolassa Nov 22 '15 at 19:51
  • 1
    Consider Detroit an object lesson. That will eventually fix itself, but it may take generations to do so. – keshlam Nov 22 '15 at 19:59

The assumption that house value appreciates 5% per year is unrealistic. Over the very long term, real house prices has stayed approximately constant. A house that is 10 years old today is 11 years old a year after, so this phenomenon of real house prices staying constant applies only to the market as a whole and not to an individual house, unless the individual house is maintained well.

One house is an extremely poorly diversified investment. What if the house you buy turns out to have a mold problem? You can lose your investment almost overnight. In contrast to this, it is extremely unlikely that the same could happen on a well-diversified stock portfolio (although it can happen on an individual stock). Thus, if non-leveraged stock portfolio has a nominal return of 8% over the long term, I would demand higher return, say 10%, from a non-leveraged investment to an individual house because of the greater risks. If you have the ability to diversify your real estate investments, a portfolio of diversified real estate investments is safer than a diversified stock portfolio, so I would demand a nominal return of 6% over the long term from such a diversified portfolio.

To decide if it's better to buy a house or to live in rental property, you need to gather all of the costs of both options (including the opportunity cost of the capital which you could otherwise invest elsewhere). The real return of buying a house instead of renting it comes from the fact that you do not need to pay rent, not from the fact that house prices tend to appreciate (which they won't do more than inflation over a very long term). For my case, I live in Finland in a special case of near-rental property where you pay 15% of the building cost when moving in (and get the 15% payment back when moving out) and then pay a monthly rent that is lower than the market rent. The property is subsidized by government-provided loans. I have calculated that for my case, living in this property makes more sense than purchasing a market-priced house, but your situation may be different.

  • The hypothetical 5% annual price increase was based on American inflation rates during the mid-to-late 1980s. At the time Lynch was writing, it was consistent with your premise that long-term housing prices tend to rise in line with general price levels. – Jasper Dec 26 '15 at 8:22
  • Ah, I forgot to consider that inflation rates used to be higher. But then again, bond yields probably were much higher back then, too, and the greater inflation rates should be seen in nominal stock yields, as well. And, the interest rates on mortgages should be higher too due to the higher inflation rate. – juhist Dec 26 '15 at 10:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.