Does borrowing help the 4 percent rule to overcome Sequence Risk?

Sequence risk is the biggest threat to a retiree hoping to live off the returns of an investment portfolio in retirement. Among the ways which have been proposed is the use of borrowing to overcome sequence risk. Whether the borrowing is done through a reverse mortgage, a HELOC, or margin finance, the aim is to avoid having to sell investments during a market downturn. While the idea is attractive in theory, as it helps to preserve the investment holdings for a future upturn in the market, testing various strategies on historical periods of sequence risk in the S&P 500 index and the Straits Times Index shows that there may only be marginal benefits of doing so.

This is the latest in a series of posts on the 4 percent rule for Financial Independence. To understand how this rule is supposed to work, what can make it fail, and whether it applies to our local context, please check out our earlier posts:

Introduction

While the idea of the 4 percent rule i.e. withdrawing 4 percent of the initial value of the investment portfolio, has been around since the 1990’s, much thought and tinkering has also gone into finding ways to minimise the impact of the Sequence of Returns Risk on retirement finances. One way which has been popularised more recently, with the innovations in finance, is the use of a Home Equity Line of Credit (HELOC) in conjunction with an investment portfolio.

The use of leverage or credit in retirement is not really something new. Traditionally, a retiree who has built up equity in his/her home can use a reverse mortgage to monetise the home equity. Under such an arrangement, the bank “loans” a portion of the value of the fully paid up freehold property to the retiree. The retiree uses this “loan” to fund his/her retirement, and finally, upon his/her passing, the bank takes possession of the house, sells it, and gets repaid. Since there can be quite a bit of uncertainty as to when the bank can take possession of the house, this really only works when the retiree has a freehold title to the house, so that the value of the house is preserved over time.

Clearly, this is not something that would be workable in Singapore, where the majority of the homes are on leasehold titles. With longevity in the 80’s for both men and women, the loss of value of the house over the leasehold of 99 years means that the open ended arrangements for repayment on the kind of reverse mortgages found elsewhere simply would not be viable for any bank.

Using leverage to make the 4 percent rule work better

But that is not the only way home equity or any sort of loan can be used for retirement. The more recent thinking on this topic is for the retiree to open a line of credit, or HELOC, just before retirement. However, the line of credit is not drawn down immediately to fund retirement. Instead, retirement income is still drawn from an investment portfolio, using a safe withdrawal rule, like the 4 percent rule. Only when the portfolio starts running out of money, or when it is battered by sequence of returns risk, is the line of credit drawn down to help tide over these periods. When the investment portfolio recovers, some of it can be sold to repay the loan. This explained in more detail here and here.

As we have shown elsewhere, selling investments in the middle of a market downturn is like dollar cost averaging in reverse, that is buying high and selling low. Perhaps a better term for it is dollar cost ravaging! So, if a line of credit can be used to preserve the investment portfolio during a market downturn, for example by using the loan to bridge the difference between the dividends from the portfolio and the amount withdrawn, the investment portfolio will be much better positioned to ride on the upturn later.

So we don’t really need a true reverse mortgage to implement this strategy – as long as a line of credit is available, whether secured on a house, or a portfolio of shares – the strategy can be used. Given the recent broad availability of margin financing, this might even be applicable to Singapore as well, although some tweaks may be needed.

For starters, you can’t actually withdraw cash directly from a margin loan secured on your shares. What you can do, however, is to sell off some of the shares a required to meet the withdrawal amount, and then buy back the same amount using the margin loan. It’s a little roundabout, but in effect, what you are doing is drawing down on a line of credit to sustain the withdrawals in a downturn without reducing the number of shares in the investment portfolio.

Does leverage really work? The case of the S&P500

But that is all in theory. Does leverage really work in practice to protect against Sequence of Returns Risk? Let’s see how it might have worked if someone in the US retired with a portfolio 100% in the S&P500 through the SPY ETF. While many of us have the impression that the S&P500 has been a great money machine in the last decade, some of us might also remember that it had a terrible decade in the 2000’s, crashing following the crash in the NASDAQ in 2000, then again following the 9/11 attack on New York in 2001, and then yet again during the Global Financial Crisis in 2008.

S&P 500 500 ETF (SPY) prices and dividends 1993 – 2020

In fact, retiring in 2000 with a portfolio 100% vested in the SPY ETF and following the 4 percent rule was probably the worst possible decision a retiree could make, beng battered by the dreaded Sequence of Returns Risk not once, but 3 times over the next 8 years! Here is how the portfolio would have ended up in 2020 after such a rough ride:

Portfolio value and withdrawals of the SPY ETF for a USD investor in the S&P 500 2000-2020

While the portfolio still could meet the withdrawals over the past 20 years, with the monthly distribution growing from $3,333 in 2000 to almost $5,000 now, the value of the portfolio has shrunk from the initial $1,000,000 to barely $500,000 in 2020. At this rate, with a required annual distribution averaging $60,000 over the next 10 years, we can expect the portfolio to run out of money within the next 10 years, even if it were all re-invested into bonds now.

By whichever metric, the period 2000 to 2020 would qualify as a legitimate case of Sequence Risk for the S&P500. How would leverage help in this case? Suppose we had three different rules for using leverage:

Borrow for 5 years: Borrow to make up the difference between dividends and withdrawal for the first 5 years only, leaving the portfolio intact. Thereafter, sell investments to fund withdrawals from year 6 onwards
Borrow for 10 years: Borrow to make up the difference between dividends and withdrawal for the first 10 years only, leaving the portfolio intact. Thereafter, sell investments to fund withdrawals from year 11 onwards
Borrow in downturns: Borrow to make up the difference between dividends and withdrawal whenever the SPY price is below the price at which the portfolio was formed in 2000. In other cases, sell investments to fund withdrawals

We will not explicitly model the repayment of the loan in each of these cases. Instead, we will let it compound at a rate of 5% per year. Also, we measure the portfolio value by subtracting the outstanding amount on the loan from the value of the shares. To make the comparison easier to follow, we break up the entire period into 2 parts: From 2000 to 2010, and from 2010 to 2020.

Portfolio value of the SPY ETF and borrowing for a USD investor in the S&P 500 2000-2010 with sequence risk

For the first 10 years, when the Sequence of Returns Risk hit the hardest, there is very little difference between all the strategies, levered and unlevered. Only in the 2008 downturn do we see that the impact of leverage is to accentuate the downturn in portfolio value compared to the unlevered 4 percent rule. So, in this case, using leverage did not help much and made it slightly worse!

Portfolio value of the SPY ETF and borrowing for a USD investor in the S&P 500 2010-2020

Continuing on into the next decade, we now see that leverage helps with the health of the portfolio. Instead of ending up in 2020 with a portfolio value of around $500,000 for the 4 percent rule, the use of a loan to protect the portfolio results in a portfolio value of up to $1,000,000! And the strategy of borrowing whenever there is a downturn in the price of the SPY works the best, followed by the strategy of borrowing for the first 10 years.

But at this point, it is helpful to consider whether it is indeed the borrowing which has helped with the outcome, or simply the rising tide of the bull run in the S&P500 in the past 6 years which has lifted all boats. Our conclusion is that it is the rising tide, and the use of borrowing did not really help that much.

Borrowing to shield the portfolio from the sequence risk of the S&P500 in 2000-2010 did not help much. Only the bull run in the late 2010’s helped to preserve the portfolio’s value

Does leverage really work? The case of the S&P500 for a Singapore investor

So borrowing didn’t really help the US investor in the S&P500 to dodge the Sequence Risk in the 2000’s. What about the Singapore investor in the same period? While it seems likely that the conclusion is the same, let’s take a look at the charts:

Portfolio value and withdrawals of the SPY ETF for a SGD investor in the S&P 500 2000-2020

The Singapore investor does even worse than the US investor! By 2020, the initial investment of $1,000,000 will have dwindled down to less than $200,000 through a combination of sequence risk, withholding tax, and exchange rate risk. Even though the portfolio is able to meet all the withdrawals plus inflation adjustments to date, there is little doubt that it will run out of money in the next few years at this rate. So, will borrowing using any of the previous three strategies help? Let’s take a look:

Portfolio value of the SPY ETF and borrowing for a SGD investor in the S&P 500 2000-2010 with sequence risk

Portfolio value of the SPY ETF and borrowing for a SGD investor in the S&P 500 2010-2020

This looks pretty much the same as the case of the US investor. Borrowing makes little difference in the first 10 years when all the sequence risk happens, but has a larger impact in the next 10 years, due to the bull run in the S&P500, and not because of any inherent benefit of leverage.

Does leverage really work? The case of the STI for a Singapore investor

So far, we see that leverage might not really work to protect the retiree from sequence of returns risk, at least not in the US or for the S&P500. What about the Singapore investor in the Straits Times Index (STI)? Now, the worst year to invest in the STI for retirement was in 2007, right before the peak (which has never been revisited yet!).

Portfolio value and withdrawals of the ES3 ETF for a SGD investor in the STI 2007-2020

Starting with an initial investment of $1,000,000, and subsequently withdrawing 4% every year, adjusted for inflation, the value of the portfolio in the STI rose to a high of $1,200,000 before plunging to $500,000 during the 2008 Global Financial Crisis. But since then, it has ridden out the sequence risk ending up at a value of around $700,000 in May 2020. One thing which the STI has in its favour (which the S&P500 does not) is a relatively high dividend yield, which reduces the need for the STI investor to sell investments when the market is in a downturn, and hence preserving the holdings in the portfolio for the next upturn.

Even as the high dividend yield of the STI reduces the sequence risk for the retiree, we still need to ask whether the possibility of borrowing against the value of the portfolio would make any difference. As before, we stick to 3 basic borrowing strategies:

Borrow for 5 years
Borrowing for 10 years
Only borrow in downturns

Portfolio value of the ES3 ETF and borrowing for a SGD investor in the STI 2007-2020 with sequence risk

Interestingly, between the 4 percent rule withdrawals, high dividend yield and borrowing, we find that borrowing does not make any difference to the outcome for the STI invested retiree.

Borrowing to shield the portfolio from the sequence risk of the STI in 2007-2020 did make a difference. It is the high dividend yield which provides a defense to sequence risk

Concluding thoughts

The use of borrowing, whether in the form of a reverse mortgage, HELOC, or even a margin loan, has been suggested as a way to protect a retiree’s portfolio against sequence risk. However, looking at the worst years to invest in the S&P500 and STI, we find that it is not really the case. In summary:

Borrowing to shield a portfolio against sequence risk does not seem to work for the S&P500 and STI
However, borrowing does help if the shares in the portfolio experience a bull run. The leverage helps to amplify the impact of the bull run
In such an instance, borrowing only in a downturn, i.e. when the price of the shares are lower than the initial investment price, works the best out of the borrowing strategies tried
In terms of sequence risk, a high dividend yield (although not mindlessly chasing penny stocks for yield) seems to work better