(originally posted here∞)


30yr mortgages have a ton of interest amortized up front. I'll bet if you add ~35% to your monthly bill, you'd have it paid off in less than 20yr and still have enough money to invest.

Play with the numbers: http://www.mtgprofessor.com/spreadsheets.htm∞

Impossible to give without knowing your goals, asset allocation plan, current investments, status of tax-deferred accounts, etc.

---

No amount of conservative strategy will turn equity-like risk into debt-like risk. For reference, Treasury bonds are yielding almost 2 full points lower than his effective ROR of mortgage prepayment on the same timeframe. That's as close to a free lunch as you're likely to find in the U.S. market.

---

Buffett put his expected long-term U.S. capital returns at 6.5% in a recent Letter to Shareholders...and that's for a guy who takes on more risk than we can (100% in equities, and most of that in value).

It's not that I don't think the U.S. stock market is incapable of high returns. Obviously it has been a great place to be for 200+ years. Two major caveats, though:

(1) Extended bull markets are fun, but they bring overpricing. Anyone planning to be a net buyer of equities over the next several years should wish that the 2000-2002 correction were bigger, because even today they're not exactly buying at a discount.


(2) It's impossible to predict which segments of the world economy will generate returns that look like our historical charts. Investing $1 in the U.S. in 1800 would've made you very rich, but who knew we'd be a better choice than Spain (the big winner of the previous two centuries)? Doing so looks wise in retrospect, but at the time was nothing more than risky (and a bit lucky). To find the equivalent choice for today's investors you'd need to take on risks along the lines of Argentina or Malaysia.

Lopping an order of magnitude off these time considerations for practicality, consider Japan 25 years ago. Things had boomed nonstop since the start of postwar reconstruction. Who knew an investor there would lose money (after inflation & taxes) for over two decades? Conversely, 25 years ago our market was wracked by stagflation. Someone chasing past returns would've likely avoided us, and thus missed the huge bull streak that prompted this discussion.

I'd surmise that young people with enough financial savvy to be considering retirement options probably aren't the ones buying new cars on credit.

---

If my phrasing implied that the figure was sans dividends, then that's my mistake (not Buffet's). Here's an example quote (not the one was thinking of, but good enough):

(source∞)

---

Dividends haven't yielded 3-4% in a couple decades, and probably won't reach that point again so long as currently-popular trends in corporate management prevail. Here's a decent summary∞.

The other Buffett figure comes from a different letter where he wrote about the shady accounting in many pension programs: by forecasting a higher expected return, companies find excuses to reduce contributions. He said he wouldn't allow any of the companies he owned to use projected stock market returns above 6.5%.

I expect my own portfolio to do better than the S&P thanks to diversification, willingness to tilt toward risk factors, and the rebalancing/correlation bonus, but even so I wouldn't put it higher than 8%.

---

...of the past. In fact, if you buy that model (I think better ones exist), then you're tacitly admitting that equities will underperform as dividends revert to their mean.

Oversimplified analogy: if you bought a 40-year bond at 4% in 1980 (corresponding to the dividend yield at the time), its asking price would have greatly appreciated as yields fell to today's levels. Yet any bond investor knows that makes it much less desireable to buy today, since the total return can only add up to 4% once the year 2020 arrives.

Stocks obviously don't have a fixed cap on future valuation like bonds, but in the simple Graham-like model under discussion they obey the same tendencies. Fama & French wrote a paper in 2001 proposing a more complex model -- I haven't read it, but from what I've gathered it largely reduces to similar principles while integrating nicely with their well-known work on risk premia.

So, overgeneralizing to illustrate the point: if stock buyback programs and stock-swap mergers and the like continue to bolster share prices, then dividend ratios will remain low. If dividend ratios rise, then equity prices will underperform. You can't have it both ways.

---

Only if you bought the market in 1980. Someone buying that hypothetical 40-year bond today will only net 1.5% total.

Anyway, this is all pretty removed from reality. Even if he picks a nice conservative number like 6%, the OP will still maximize his future net worth by putting excess income into a diversified portfolio. Given that makes some huge assumptions -- e.g. he will have the discipline to write his brokerage a check every month, whereas his mortgage lender is a bit harder to ignore -- it makes sense to pursue a balanced strategy like TomM's. Play with the spreadsheet I linked in my first reply; I'm sure you'll find a solution that works for you.

---

One should have an emergency fund established before considering any of the proposals in this thread.

Your post is a good reminder, though: someone with a lot of equity in their home can use a HEL-backed line of credit as an emergency fund. No interest so long as you don't withdraw anything; meanwhile, you can move most of your savings account to something with less liquidity but higher yield, e.g. I-Bonds.

---

Back to CollectedWritings