The TIPS-Dividend Approximation

Here are two posts that help us decide whether we should pursue a dividend-based strategy or a stock growth strategy during retirement. I wrote the first part on 8-18-04.

TIPS and dividends share an important characteristic. They both produce steady streams of income. This is in striking contrast to the stock market as a whole. The most widely known investigations involving retirement portfolios have made us acutely aware of the damage caused by price volatility. Selling when prices are low destroys retirement portfolios. To maintain safety, we must reduce the amount of income that we withdraw. Dividend-based strategies can allow us to withdraw more because the income is steady. There is no need to sell in a down market.

Our calculators do not allow us to examine most dividend-based strategies directly. But we can take advantage of the similarities between TIPS and dividends to examine dividend-based strategies indirectly.

What I have done is to look at the 30-year balances of pairs of portfolios of stocks and TIPS. Each pair consists of a 20% TIPS / 80% stock portfolio and an 80% TIPS / 20% stock portfolio. The high-TIPS portfolio provides a steady income at its (real) interest rate. This is similar to a dividend-based income stream with the same yield, provided that it grows enough to match inflation. Even if it fails to do so, it is likely to behave in a manner more readily acceptable to retirees than alternatives. Any shortfall in matching inflation is because dividends have failed to catch up. It is not because of selling shares. Financial adjustments are natural.

There is a crossover point. At some interest rate, requirements for growth favor owning more stocks. At another interest rate, requirements for safety favor owning more TIPS.

Crossover occurs when a high stock portfolio and a high TIPS portfolio produce comparable results. We make our comparisons by requiring a high level of safety. For reasons identified in previous investigations, I limited my analysis to 1923-1975.

General Conditions

My procedure was to look at 30-year portfolio balances using withdrawal rates of 3%, 4% and 5% of the initial balance with adjustments to match inflation. I looked at TIPS interest (real) rates of 0%, 2%, 4% and 6% initially. Later, I took data at 2.5% and 3.0%.

I set the initial balance to $100000. I set expenses at 0.20%. I included annual rebalancing. I reinvested all dividends and interest. I did not remove a fixed percentage of portfolio gains. [Withdrawals are made separately. They are applied to the entire balance, not to individual components.] I used the CPI for inflation adjustments.

You can duplicate these results on the Retire Early Safe Withdrawal Calculator, version 1.61 date November 7, 2002 or any of my modified versions. The 30-year balances are in column AB, rows 15-124.

Analysis

I made plots of the 30-year balances for each pair of allocations (20% and 80% TIPS) versus earnings yield. As might be expected, the 30-year balances with 80% TIPS grew slowly from their base and showed very little scatter. Portfolios with 80% stocks and 20% TIPS had considerable growth and a large amount of scatter.

Focusing on safety, I noticed that the high stock portfolio failed once [in 1966] when the withdrawal rate was 4% and TIPS had a 2% interest rate. The low stock portfolio did not fail under those same conditions.

[Do not be surprised that TIPS at 2% interest are safe at a 4% withdrawal rate. Remember that TIPS at a 0% real interest rate allow you to withdraw 3.33% safely every year for 30 years. The final balance is zero.]

Taking a more general view, the high TIPS portfolios were much better than the high stock portfolios when interest rates were 4% and higher. The high stock portfolios were better most of the time when the TIPS interest rate was 2% or lower.

Looking at 2.5% and 3.0% interest rates, preferences were a toss up. The Great Depression favored stocks while the 1960s favored TIPS. All in all, I consider interest rates of 2.5% to 3.0% to be a transition region (when looking at 3%, 4% and 5% withdrawal rates).

Conclusions

TIPS become more attractive than stocks when interest rates are 3.0% and higher. They are less attractive when interest rates are 2.5% and lower.

The crossover point is 2.5% to 3.0%.

Here is the TIPS-Dividend Approximation: At high levels of safety, a dividend strategy is better than a high stock strategy if it can provide an initial yield of 2.5% to 3.0% and grow enough to keep up with inflation.

Additional Remarks

What we have established are the requirements for a dividend-based strategy to become more attractive than an overall stock market strategy. Dividend yields need to be 3% or more assuming that their growth matches inflation. These requirements apply broadly for withdrawal rates of 3% to 5% at the 30-year point. Once these requirements are met, one can focus on factors other than safety.

These requirements apply to TIPS as well. Whenever yields equal 3% or more, a portfolio should be weighted heavily in favor of TIPS. When yields are between 2.5% and 3.0%, the weighting matters little. When yields are less than 2.5%, stocks become more attractive.

[These values include conditions that fail. That is, the dividend-based strategy has a similar likelihood of surviving or failing when yields are 2.5% to 3.0% as an overall stock market strategy. If the portfolio survives, the balances are similar at 30 years.]

Interestingly, a portfolio that emphasizes dividend income can have a better upside than one emphasizing the overall stock market. The reason is that the return from the holding other than stocks, especially if it is commercial paper, can drag the return of the total portfolio well below that of stocks. A dividend-based strategy is likely to come very close to matching the high, long-term return of the stock market because it can tolerate a very high percentage of stocks. If it were not for considerations such as liquidity and the redeployment of assets, the stock allocation could be 100%.

There is a higher potential upside with a dividend-based strategy than with TIPS. However, the safety falls below 100% and there is a much greater possibility of loss if some shares have to be sold.

The required dividend yield of 3% is modest. There are many quality stocks that meet this hurdle. The requirement for enough growth to match inflation is not too difficult either.

Now let us look at somewhat different dividend yields.

Consider a stock with a dividend yield of 4%, but without any growth. We start with the formula for compounded returns: [final balance / initial balance] = (1+r)^n, where r is the interest rate and n is the number of years. We adapt it to see what it takes for $3 to grow to $4 at the rate of inflation. If inflation is 3%, the equation becomes [4/3] = 1.03^n and n is 9.7 years. If inflation is 4%, the equation becomes [4/3] = 1.04^n and n is 7.3 years.

This tells us that such a stock exceeds our requirements for 7 to 10 years provided that inflation is 3% to 4%. It would take that long for dividends beginning at 3% to grow to the same dollar amount as initially thrown off by a stock yielding 4%. The surplus could buy additional shares of stock as well. Moreover, 7 to 10 years is a very long time for us to make any dividend projections.

Now look at a stock that begins at a 2% dividend yield. If its dividends grow by 10% per year, they will double in 7.3 years. [For those using the rule of 72, the exact value of n is 7.2725409, which is very close to 72/10.] Notice that this is the same amount of time that it takes $3 to grow to $4 at an inflation rate of 4%.

This tells us that such a stock would need to grow its dividend by 10% or more for 7 years to catch up to requirements. It would be throwing off less than the required dividend amount throughout the entire amount of time. Even if the dividend growth rate were 20%, it would still take 3.8 years to catch up.

Have fun.

John Walter Russell