By: Dividend Sensei
Last week the market had its worst week in two years, and on Monday February 5th, the S&P 5oo had its worst day in seven years. Then this morning (February 6th), the S&P 500 officially hit -10% off all time highs and entered a correction. Naturally many investors are shocked at the severity and speed of the drops (we had a flash crash yesterday). It’s important to remember that corrections are a natural, and healthy part of a bull market. In fact, corrections are necessary in order for overvalued markets to get in line with fundamentals (that actually determine long-term intrinsic value) and thus allow bull markets to continue for long stretches of time.
For example, before this correction, Morningstar estimated that, among the stocks it analyzes, the average overvaluation was about 11%. Today stocks are just 2% overvalued, and with earnings expected to grow 18% this year (thanks to tax cuts) this means the market has the potential to rise 16% off today’s levels and end 2018 up 8%. While that may be a far cry from last year’s 20% increase, it’s still a decent return in what is now the second longest bull market since World War II.
Of course, a positive 2018 is all well and good, but what everyone wants to know right now is “how bad will this get” and “how long will it last”. No one can predict short-term prices, so I’m not going to foolishly predict a fixed end date, or exact point level that the S&P 500 will bottom at and then take off to hit new all time highs. But what we can do is analyze previous corrections to see what the average duration was, and how severe they were.
According to Deutsche Bank since 1957 corrections occur once every 357 days on average, or about once every 1.5 calendar years. So this one is actually slightly overdue (last correction ended Feb 11th, 2016). As for duration, and severity? According to John Prestbo at MarketWatch, between 1945 and 2013, the Dow Jones Industrial Average’s mean correction was -13.3% and lasted 71.6 market days, or about 14 weeks. Note that the duration is measured from the last market top (January 25th, 2018), and the ultimate bottom for the market.
Of course, the Dow Jones is hardly representative of the broader market, since it’s not even an index at all. Rather it’s a pure average of 30 somewhat arbitrary stock prices. And as we saw over the past year, it can be more volatile than the S&P 500, which itself can be non representative of various stocks that don’t contribute to its index (S&P 500 is a market cap weighted index of 505 of America’s largest companies). Basically this means that the S&P 500 may not be set to drop as much as the Dow, perhaps only 11% or 12% or about 3% to 4% more than today’s level.
That actually fits nicely with Bank Of America’s recent report that indicated that it expects the S&P 500 to potentially decline by about 12% within three months of January 24th (when it issued its report triggered by historic market inflows of cash). However, it’s important to realize that the duration of any correction is ultimately made up of two key factors. First, the aggregate macro economic, and corporate earnings picture, and the severity of the crash itself.
For example, corrections that took place in 2010, 2011, 2012, and 2013 were all triggered by worries about slowing global growth, but especially a weak US economy and the potential for these to harm corporate earnings growth. Today the situation is completely reversed. In fact, the market is now worried that the economy (and wage growth in particular) might be too strong. Specifically there is a concern that too much business investment (partially triggered by tax cuts) might cause wages to rise fast enough to raise inflation and force the Federal Reserve to raise interest rates four times in 2018 (current plan is 3).
Now of course the Federal Reserve’s Fed Funds rate doesn’t directly affect anything other than the overnight interbank lending rate. However, banks do benchmark their prime rates off this, and the prime rate ultimately is what determines consumer spending costs (personal loans, credit cards, auto loan rates, etc). Theoretically if the Fed hikes rates too fast than deeply indebted American consumers might face rising interest costs that force them to cut back spending or even default on their debts. This could then hurt true economic growth and potentially even cause a recession (but one that’s far milder than 2008-2009).
What really spooked Wall Street however, was long-term interest rates, which are set by the $14 trillion US Treasury bond market. Bond yields are market driven, with prices based on three things: coupon (that bond’s interest rate), duration to maturity, and long-term inflation expectations. In the past quarter the market’s long-term inflation projections have risen from about 1.8% to 2.1%.
This has caused 10 and 30 year Treasury bonds (the most inflation sensitive) to sell off in a truly impressive display of one of the fastest bond meltdowns in years. 10 year Treasury yields (proxy for long-term interest rates that affect corporate borrowing costs) rose from 2.46% at the start of the year to a max of 2.88% on Monday morning. A 0.42% increase doesn’t sound like much, but in the usually slow moving (and enormous) bond market, it’s lightning quick.
This rapid rise in yields is what caused dividend stocks, especially high-yield sectors like utilities, and REITs, to suffer during the broader market’s best single month since March 2016 (S&P rose almost 6% in January). However, finally long-term rates got so high that suddenly investors began to question whether they might end up slowing economic (and earnings growth) and so we got Friday, February 2nd’s 2.5% market drop.
That seems to have caused a chain reaction, potentially spurred on by massive passive investor (ETF owners) selling in fear, and then machine trading bots made it worse via blindly adding to the fear (via Monday’s flash crash). Ironically enough the market panic has driven investors to bonds (flight to safety), and so 10 year yields are now down to 2.75%. That’s still high by recent standards, but it means that the very thing that caused this correction is now gone.
The bottom line is that this correction doesn’t reflect anything fundamentally wrong with the US economy, or corporate earnings growth. Thus it’s a “technical” correction and not a “fundamental correction.” That means that it’s unlikely that we fall much beyond the 12% to 13% that is the market average over the past 80 years or so. And as for the duration? well remember that because there is likely a fundamental cap to how low stocks can drop (economy is just too good to likely all for a bear market right now), this means that the faster the market falls to -12% to -13%, the faster we’ll bottom.
And since the recovery from a correction is usually very quick (once the panic and uncertainty are gone), the month following the bottom of the market is usually the strongest month for stocks for several years. In fact, usually the strongest single market days happen within 2 weeks of the worst market days. This means that, historically speaking, we probably only have 3% to 5% more to drop. Whether that takes 2 days, 2 weeks, or 2 months, just remember that corrections are the single best time to put long-term money to work in quality investments (like high-yield dividend growth stocks).
So don’t let the recent interest rate spike induced correction scare you, because ultimately short-term share prices are vanity, but long-term cash flow (and dividend) growth is sanity.