Author Topic: FIRECalc Gurus - What is the impact of borrowing $250k to invest in the market?  (Read 1859 times)

Cycling Stache

  • Bristles
  • ***
  • Posts: 471
  • Age: 43
This came up in another thread in a discussion about rational investing, and I realized that some of you are really good with FIRECalc and cFIREsim and could give actually mathematical answers.

Most people agree that pre-paying a low-interest mortgage rather than investing is rationally incorrect.  Over the long term, the market should return 7%, and a 3% interest loan is worth the cost in order to invest.  It's easy to see that's correct over the typical 30-year mortgage period because the numbers are significantly higher for the market returns.

But there are also seem to be very few people who actually take the equity out of their house in the form of a HELOC to invest, even though the concept is exactly the same. 

The latter seems scarier, even though we understand (or should understand) that rationally it's no different.  And yet, our reliance on the 4% rule and investment portfolios to sustain early retirement is premised on being (at least relatively) risk neutral and trusting the numbers.

So what do the FIRECalc or cFIREsim numbers show about borrowing money to invest in the market?

If you have $750k in investments and can borrow $250k in home equity at a fixed 3% with a 20-year repayment period, does that always increase your odds of success?  What kind of increases?  Assuming the interest rate is locked and you're never going to get the equivalent of a margin call (i.e., being forced to pre-pay the loan or conduct a fire sale--excuse the pun), is there any situation in which borrowing the home equity doesn't make rational sense in the calculations?

I realized that, even though I consider myself a fairly rational investor, I'm not being rational by both (1) counting on the 4% rule, which requires a risk neutral acceptance of the numbers, and (2) refusing to consider investing home equity, because I don't like having debt.  I'm now curious what the numbers show.

P.S.  There was an example given in the other thread about a Boglehead investor who tried to borrow $200k in 2007 to invest early in his career, and it all melted down when a series of margin calls in late 2008 wiped him out.  The quick reaction was don't ever do it, citing that example, but interestingly, that was a historically bad time to invest, and really not any different than citing the 5% failure rate as a reason not to follow the 4% rule.  And even more interestingly, even though that was one of the worst times ever to invest, if he had not gotten the margin call, he probably would have made money on the loan through today, assuming a 4% interest rate or so on the loan.

Curious to see how this turns out.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
a HELOC can be called on at any time for starters so its way scarier.  your post should be for an HEL.  or just a REFI to a longer term. 

if you assume a 250k house and you have 750k in investments this is really easy to input into cfiresim you dont have to be good at math.

scenario 1 paid off house spending is 4% of 750k or 30k per year 75/25 stock bond  40 year scenario for money to stay alive
89.62% chance of success

scenario 2 assume house not paid off but 1MM invested and annual spending at 30k not including mortgage, mortgage payment based on 20 years at 3% - 16632 - same AA and life expectancy needed
91.43% chance of success

but at 30 years and a 3.25% mortgage you have a
94.56% chance of success

you just make the total portfolio 1MM and add the 16632 to the bottom as a non inflation adjusted cost

why is the 30 so much better? b/c the market has performed much better over 30 year periods on avg than 20 year. 



Cycling Stache

  • Bristles
  • ***
  • Posts: 471
  • Age: 43
a HELOC can be called on at any time for starters so its way scarier.  your post should be for an HEL.  or just a REFI to a longer term. 

if you assume a 250k house and you have 750k in investments this is really easy to input into cfiresim you dont have to be good at math.

scenario 1 paid off house spending is 4% of 750k or 30k per year 75/25 stock bond  40 year scenario for money to stay alive
89.62% chance of success

scenario 2 assume house not paid off but 1MM invested and annual spending at 30k not including mortgage, mortgage payment based on 20 years at 3% - 16632 - same AA and life expectancy needed
91.43% chance of success

but at 30 years and a 3.25% mortgage you have a
94.56% chance of success

you just make the total portfolio 1MM and add the 16632 to the bottom as a non inflation adjusted cost

why is the 30 so much better? b/c the market has performed much better over 30 year periods on avg than 20 year.

Okay, this is helpful.  Why do you say the HELOC is callable?  I was under the impression that once you borrow the money, you're locked into the pre-payment period set by the contract, assuming no catastrophic changes (in my case, even a large loan is not going to offset the value of the house)?

Is there any set of circumstances short of reducing the time frame below 20 years that would make this a negative outcome?  I learned how to type in the basic assumptions and modify them, but I didn't know whether there's some way to confirm that a certain type of change would always produce a certain type of outcome.

Does a $250k loan payable over 20 or 30 years always produce a higher success rate, no matter whether you modify the starting portfolio size or the asset allocation (stocks/bonds)?  Is there a way to figure it out without just repeatedly typing in different assumptions?

And if the $250k loan is ALWAYS a more successful outcome, does that mean that there is no mathematical reason (at least based on the historical data) not to take it so long as the interest rate is locked in and it can't be suddenly recalled? 

(I'm not sure how essential the locked in interest rate is, although I'm betting it helps.  Still, the market typically returns a premium over inflation, so you would expect to see a long-term positive return over the prime rate regardless of what that prime rate is).

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
you realize cfiresim is just testing your inputs historically so.  yes i would assume there is an asset allocation that could be worse than a non fixed rate loan ... but you cant really input a variable rate loan into cfiresim.  i mean whats your end game here. 

Cycling Stache

  • Bristles
  • ***
  • Posts: 471
  • Age: 43
you realize cfiresim is just testing your inputs historically so.  yes i would assume there is an asset allocation that could be worse than a non fixed rate loan ... but you cant really input a variable rate loan into cfiresim.  i mean whats your end game here.

I believe taking a home equity loan to invest in the market is conceptually no different than electing to invest rather than pre-pay mortgage.  I'm curious whether the historical data bears that out.  And, specifically, whether the calculators will tell us whether there is always a greater success rate when borrowing e.g., $250k at a fixed rate versus not borrowing.  I can type in a few examples to see how it works, but I was hoping someone better at using the calculators could figure out how to confirm that. 

If it's always a positive outcome, what can be the rational argument against taking out the home equity loan and investing it?  Seeing the historical success rate I believe would be helpful.  If it occasionally is worse, what situation is that?

I also caveat it by assuming that a person already has a sizable stash built up, so that it's not failing because of the immediate need to pay money back on the loan and no assets to draw on if the market is down.  It would be neat to see those calculations borne out if so, because I believe there is a lot of emotional resistance to borrowing money to invest that could be shown to be mathematically incorrect. 

Assuming the calculations show that, of course.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
so run those calculations and learn to use the tool but over all 20 and 30 year periods it will be better.  and as stated originally most HELOCs can be called at any time.  so you'd need a HEL.  not a HELOC ... HELOC's also tend to be variable.  the current rate climate has only existed in recent history post 2008.  so these rates are not normal and you will not always be able to take a HEL out at these rates.  there are also costs with refinances and HEL's etc. that need to be taken into account. 

the rational arguement against would be sequence of return risks.  these are increased when you have a fixed rate loan in the event the us market does something it has never done before and does what japan has done and remains flat.  again this is just against historical data and the arguement against would be that just b/c its never happened does mean it wont. 

so to sum it up if you

 have a low fixed rate loan that cant be called at 20-30 year term you're probably fine.  why would you want to figure out where the worst case scenario/breakeven was historically. find the best case scenario enable that or dont and move on

BTDretire

  • Magnum Stache
  • ******
  • Posts: 2625
I already know about market timeing, BUT!
How old are you?
Do you have many years to recover from a market loss?
 I was around to see the Nasdaq drop 76% in about 18 months starting Mar,2000.
I also watched a 51% market drop in 2008.
The Nasdaq is up 377% since Feb, 2009.
Is this a good time to pull equity from your home?
I don't know, but I'm not doing it.
I'd love to see 8 more years of stock market growth, but I'm
being a bit more conservative, with maybe $100K in cash.
I don't even want bonds, in this low interest rate environment.
 My crystal ball is just as good as anyone elses and nobody elses is any good.
Good luck.

Edit, I see I didn't answer your question.
That's OK.