The main argument of Getting Back to Business is that investors should approach portfolio construction and management from the perspective of tangible cashflows distributable and distributed to the end investor. For the equity portion of one’s portfolio, that generally means leaning in the direction of dividend-paying stocks and higher-yielding ones at that. But wait, own dividend stocks in the current rising interest rate environment? “That’s crazy!” say many market participants. “You are going to get crushed!” they warn. The non-dividend part of the market will do much better, continues the argument. That’s the accepted wisdom. With interest rates having recently reached multi-year highs, I want to unpack that notion and see whether the accepted wisdom actually withstands logical scrutiny. I think not, but you can be the judge.
First of all, let’s dispense with the easy stuff. Rising rates on their own, particularly near-term moves or increases that just get us back to more normal levels, as we are seeing currently, simply have little to no bearing on the cash generation of most high-quality, dividend-paying companies. Yes, rising rates may have a modest impact on working capital and other real-world business activities, but they are limited in scope, and remain far less important to a company’s cash generation than higher-order changes in sales or margins. Of course, with rates rising, the cost of long-term debt goes up, but here too the change will be slow, and it will show up on the income statement over many years as existing debt comes due and gets refinanced at higher rates. Keep in mind that investment-grade companies currently re-financing 5 or 10 year-term paper are generally doing so at lower rates than when the debt was issued. That is, debt refinancing can still be an income-statement tailwind for debt-capitalized companies. That will come to an end soon, but let’s be clear that having rates return to more typical levels is in no way an operational issue for most large, dividend-paying companies. Should rates move to very high levels—think 1970s and 1980s levels—the story changes. Extremely high rates can be disruptive to businesses, especially those without strong pricing power, but we are simply nowhere near those levels.
That being said, rising interest rates can and will eventually have a dampening effect on the overall level of economic activity, particularly for business models—including far too many consumers—that are dependent on cheap credit for everyday activities. If a recession or even just a general economic slowdown were to result from the current cycle of rate hikes, that could affect the revenue and profit growth of all companies, and their prospective dividend payments. While that is a valid concern, and the stock market as a discounting mechanism takes that (and other) concerns into account, the simple trader’s mantra—rates up, dividend-paying stocks down—is absurd when considered from a distributable cashflow or business perspective. And keep in mind that long-term interest rates still remain at exceptionally low levels by historical standards. If some business begins to struggle because the US 10-year Treasury is now at 3% rather than 1.5%, one should ask how robust its business model was to begin with. (We won’t even deal with highly leveraged companies whose whole existence was predicated on the existence of super cheap money. They will undoubtedly struggle in an existential way in a rising rate environment.)
OK, but what about share prices. “These bond proxies will get killed when rates rise.” I have heard that more or less every day for the past decade, and market data suggest that, at least near-term, there is a hint of truth to it. But successful long-term investors need to move beyond hints and understand fundamental truths. Just because dividend-oriented stocks may trade off on a day the 10-Year US Treasury moves up 5 basis points doesn’t make that a logical outcome or the basis of a long-term investment strategy. Traders justify the lower bids by arguing that those stocks are like bonds and that their net present value, and therefore their price for purchase, should adjust downward when rates of interest, the basis for a discount rate applied to the bond payments, move up. In the case of traditional bonds, that logic may be reasonable. Typical bonds pay a certain fixed coupon amount of interest twice a year and then repay the principal at maturity. Increase the rate at which one discounts those payments during the life of the bond and the price of that security and the right to its future income stream adjusts downward. For a stock whose dividend is flat and not likely to rise anytime soon, the logic is largely the same.
But for the vast majority of dividend-paying companies, the logic is not the same. One ordinarily owns these types of stocks for their increasing dividends, not a flat payment stream. Stocks represent an ownership stake in the underlying business, and the opportunity to grow along with it. As sales and profits increase, so too do the dividends, all other things kept equal. Interest rates, among other functions, are a reflection of inflation and economic growth expectations. If rates move up, it is because inflation or real GDP is gaining or expected to soon. At the macro-economic level and for most large companies operating at that macro-economic level, that means nominal sales and nominal profits will also be moving up. There is a bit of cart and horse here, and there can be lags, but rising rates should generally be met with rising growth in dividend payments. In short, dividend paying stocks are not bond proxies and should not be treated as such. If the profit and dividend growth rate can “keep up” (we’ll get to that in a moment), then there is no reason why the Net Present Value (NPV) or the fair price of dividend-paying equities should suffer along with their flat-payment bond cousins.
Now I am well aware that exactly zero market makers or algorithmic traders are engaged in this type of calculation when they are making their bids, but it does suggest a basic market inefficiency. In fact, the knee-jerk “sell the dividend payers” creates a nice if modest investment opportunity. If share prices for a basket of dividend-oriented equities drop 5% due to an increase in the 10-Year that has essentially no impact on the ability of those companies to pay and increase their dividends in the future, then the savvy investor with either new cash to invest or dividends to re-invest gets the same companies and the same income streams for 5% less in cost (or a 5% increase in yield.) More generally, investment strategies based on a tangible component of return—cash—offer that unique benefit: a temporary drop in the share price increases the expectable cash return by the same amount from that point on, as reflected in the higher dividend yield. Well done.
The only real challenge to high quality, dividend-growing equities from near-term rising rates is not in the dividends themselves or even the math of present value, but in the simpler notion of alternative sources of income that could reasonably affect the demand for dividend-paying equities. In a rising rate environment, cash, money-market funds, government securities, or corporate bonds will eventually begin to offer yields that are attractive to some investors, especially those with short time horizons, no need for a rising income stream, and no particular love for the stock market. For them, moving to those instruments may make sense. But it is folly to imagine a massive shift in investor allocations that would justify a sharp sell off in dividend-paying equities in favor of other income streams. There just aren’t enough Grannies with three-year time horizons to justify that kind of move every time rates move up 5 or 10 basis points. With seniors living longer and longer, the opposite is true. A rising income stream holds a certain allure that perhaps partially offsets the discomfort that seniors can have being in equities.
Now on to more serious issues. If rising rates have no material cashflow impact on high-quality dividend-paying equities, investors will still ask how rising rates might affect how one thinks about the cashflow-based approach to portfolio construction and management advocated in Getting Back to Business. It’s a fair question. From a cashflow perspective, the portfolio’s yield (or yield at cost) and the expected distribution growth rate are the single most important factors that investors should keep in mind (as well as the volatility of the actual income stream), but it is useful to provide investors with an additional tool that more directly addresses the impact of a changing interest or risk-rate structure.
Here is where I think stockmarket investors could benefit from an analytical framework used in the fixed-income world. It’s called “duration,” and it is used to measure the sensitivity of a bond (or a portfolio of them) to a 100-basis point change in interest rates. The underlying concept of duration was created in 1938 by Canadian economist Frederick Macaulay. His original formula, now known as Macaulay duration, comes up with the length of time a bond investor has to wait to receive the net present value of the cashflows from the bond or a portfolio of bonds. It is a measure of time. The higher the number, the longer it takes you to “get your money back”. When you then layer in a change in interest rates used to discount those cash flows, you have a back-of-the-envelope tool to measure the change in the “proper” price of said bond or portfolios. For simple bonds, the higher the duration, the greater the impact of an interest rate change on the bond price because there is more time for the discounting process to impact the present value of the cashflows. The lower the duration, the less the window of opportunity for a change in interest rates to have on the price. If we assume that interest rates are likely to rise over the investment time horizon of the bonds—and we naively keep all other factors and considerations equal—investors would prefer a portfolio with a lower duration rather than a higher one. The rising rates, would, in theory, do less “damage” to the price of the lower-duration portfolio. A shorter maturity is the easiest way to have a lower duration. A notably higher coupon amount, or yield, for any given maturity, is another way to generate a marginally lower duration. (This summary of bond duration will most certainly not get you through your CFA® Level II Exam, but it is sufficient for our purposes.)
In Getting Back to Business, I argue in favor of constructing and measuring equity portfolios in light of their cashflows to the end investor. Applying the concept of duration to equities would then constitute an additional tool to see how this portfolio would perform—in terms of near-term price movement—in a rising-rate environment compared to other types of equity portfolios. In a traditional duration calculation, most of the variables are fixed—you know the current price of the bond, the amount of the coupons and the principal, and the current interest rate. Change the interest rate and calculate. For equities, the math is a lot harder. Equity coupons—the dividends—generally grow over time, so you need to calculate that into the equation. And then there is the matter of the terminal value. With simple bonds, you have a date of maturity when your principal is returned. That principal value gets discounted back to the present time. For equities, there is no date of maturity. So either you calculate the duration of the income stream out in perpetuity, which is tricky, or you come up with a terminal value of the stock some time out, and discount it back to the present time, which is trickier. Finally, there is the matter of the right discount rate to use. The widely taught CAPM model for an expected return—the same as a discount rate—is one option, though, as argued in Getting Back to Business, it is not a very good one. But regardless of approach, you still need to come up with a discount rate as part of the duration calculation.
The formulas get even messier because of the interplay of the variables. For instance, there is presumed to be some correlation between growth and inflation, particularly in a “nominal” price environment. That is, you pay for your softdrink and your monthly phone bill in nominal, not inflation adjusted, dollars. And the dividends that you get from your portfolio are also in nominal dollars. So some of the change in the growth rate of those distributions and the discount rate used in the duration calculation would be related to one another. The question is by how much are they related. The number and interplay of variables is such that there is no standard, widely accepted and straightforward way to calculate the duration of equities. This question is still waiting its “CAPM” solution; in the meantime, the numerous back-of-the-envelope approaches that are available will have to do.
And then there are the really tough questions. However calculated, the duration exercise is designed to determine the asset-price sensitivity to a change in interest rates, or in the case of equities, a change in the discount rate for the cashflows. But what about a rise in “real” rates, those not directly related to changes in inflation? If rates rise without inflationary pressures—a real risk in that policy makers are trying to get rates back to “normal” levels after a decade of quantitative easing—you have yet another factor to deal with, and one with perhaps the greatest potential impact on prices. That is because current real risk rates—using the 10 Year Treasury as a base—have been at or close to zero (or even negative at times) for the past 7 years. (Simple math: 10 Year Treasury during that time at 1.5% to 3%, inflation at that time around 2%. Real risk rates around -0.5% to 1%. Very, very low.) The traditional discount rate for equity cashflows is also based off the 10-year “risk-free” rate with an additional equity risk premium that has been around 5% nominal in recent decades. If you inject “real” risk back into the equation—either at the risk-free rate or via a higher equity risk premium, you open up a whole new series of outcomes.
But even without doing all the math, certain concepts stand out. A stock portfolio offering current material cashflows (dividends) will necessarily have a lower duration than one with no or low current cashflows. That’s for the simple reason that there are “upfront” payments which get reduced less by an increase in rates. Think about that when you consider the standard notion that growth stocks will do better in a rising rate environment because they will “keep up” with inflation better. Maybe yes, maybe no, but without an income stream (or just a minimal one) that rises along with the inflation, the math of duration suggests otherwise. In the Strategic Dividend Investor (from 2011), I cite an example that helps make the point about the risk of deferring cash returns decades out in to the future. If you compare the income stream from a portfolio yielding 5% and growing the dividend at a rate of 5% annually with an income stream yielding 2% and growing the distribution 8% per year, both portfolios have the same expected annual return of 10%. But despite the higher growth of the 2 + 8 portfolio, it takes a whopping 33 years for the annual income of that portfolio to equal the annual payments from the 5 + 5 portfolio. (page 57-58). This is not a duration calculation, but just a common-sense glimpse at cashflow management. And it should not come as a surprise that a rise in the discount rate will have a far greater impact on the “growth” portfolio than on the one offering the higher current cash distribution.
Defenders of the low or no-cashflow approach are put in a tricky position if they want to pay nominal obeisance to the discounted cashflow framework which still underpins most valuation exercises on Wall Street, even for non-dividend paying stocks. Their only real answer is that once their distributable cashflow-free companies have grown to a certain scale and profitability, they will begin to make distributions of such size and growth rates as to justify the many “upfront” years of no or low payments. In Getting Back to Business, I used the example of a leading on-line retailer whose cash distributions may start in a year, in a decade, or never. (See pages 250-252). The math of duration, however, is unequivocal: Those long-deferred payments, no matter how much is promised, will be worth a lot less today if discount rates move up, and especially if real rates increase materially.
So I’m going against the crowd here, but the logic of a discounting mechanism is neither popular nor unpopular. It just is. Cash is cash. And business investors with a cash sensibility and the ability to see through near-term share price moves can turn the current mania about rising rates into an opportunity. Obviously, holders of non-cashflow securities and near-term traders care not a whit about these calculations, at least for now. We will see if that continues if real risk, not just inflation-linked risk, reenters the investment equation a decade after it was abolished by central banking authorities.
For those investors who do care about cashflows, there are ways to at least partially “immunize” their equity portfolios against the prospect of rising rates, whether due to inflation or the addition of real risk. They include:
- Where you can, tap into cashflows now rather than in the future. That means a higher rather than lower current yield, as high as can be had while still maintaining a rising, high-quality income stream able to offset inflation.
- Own companies with as much pricing power as possible, again to offset the impact of inflation. (This can turn out to be a challenge as the best dividend payers are, in the current environment, in the Old Economy, where pricing power can be hard to come by. That’s where active management and fundamental analysis come in.)
- Be very careful about your exposure to rising real rates, that is, the return of genuine investment risk. If rates move up more than inflation, it will likely be tough on “at risk” securities. Make sure you are aware of which investments—highly cyclical, highly leveraged, thin business models, etc—fall into that category and monitor them closely. If real risk reenters the investment calculus after a decade-long hiatus, you will want to lean toward less-cyclical, well capitalized businesses. They will suffer less.
There’s lots more on this and other related issues in Getting Back to Business.
Daniel Peris October 31, 2018.
Note: The views expressed here are those of the author alone, and do not necessarily reflect the views of his employer. Nothing written here should be construed as investment advice. Consult your investment advisor for specific recommendations.