By Chris Metli, Executive Director in Morgan Stanley's Institutional Equity Division
Buying the dip has been the right trade for so long that it has become ingrained in investors’ psyche. Ample liquidity in the QE era provided the ammunition, and with robust earnings growth investors learned that selling into a downturn was the wrong trade, particularly during 2016 and 2017 (think the Brexit and US election selloffs and sharp rebounds). Even corporates have learned to buy the dips.
The problem is that the world is changing, and the environment is shifting from one where dips should be bought to one where rallies should be sold. 1) QE has ended and global central banks are withdrawing liquidity, 2) earnings growth is set to slow from recent tax-cut fueled highs, and 3) poor performance and volatility is limiting risk appetite – as portfolios are whipsawed investors will slowly learn that dips are no longer meant to be bought.
Price action in 2018 already shows that ‘buy the dip’ is on its way out. Buying the S&P 500 after a down week was a profitable strategy from 2005 through 2017, and buying these dips fueled most of the post-crisis S&P 500 gains (relative to buying after the market rallied). But in 2018 ‘buying the dip’ has been a negative return strategy for the first time in 13 years.
This shift is occurring in part because central banks are withdrawing liquidity and hiking rates, meaning there is less cheap money available to buy assets – both for investors and for corporates. And now this tightening in financial conditions is occurring as strength in earnings growth is starting to be questioned.
But the dynamic that will ultimately convince investors to stop buying the dip is performance. P/L for active managers (both HFs and MFs) is poor, which limits risk appetite, particularly at this time of the year. While some investors have tried to buy the dips this month, the market has not been able to hold onto rallies (nor can single names that beat, i.e. NFLX). This sends a signal that demand is not what it used to be, and as investors lose money buying dips their behavior will change.
So far, investors have been managing risk in this selloff by putting on shorts while not necessarily selling their core longs (i.e. the crowded Tech and Growth names). The pain may not stop until the ‘rolling bear market’ as MS Equity Strategist Mike Wilson notes (see Dead Cat Bounce; Risk Reward Remains Unattractive for US Equities; Risk Premium Q&A Oct 22 2018) takes out the final holdouts, and forces a proper de-grossing of those positions.
The deterioration in ‘buy the dip’ also extends to shorter time horizons as well, with the market seeing relatively larger ‘gaps’ and momentum during the trading day. The below chart shows how S&P 500 futures prices are either mean reverting or trending intraday (it plots the correlation of front futures prices one second to the next, with negatives indicating more mean reversion and positives indicating more intraday momentum, or ‘gaps’ in price). There is a general trend higher in momentum / towards less mean reversion over time. And importantly on each volatility spike there is relatively more momentum and gap risk intraday – February 2018 looked a lot like 2008 on this metric, despite the VIX only rising half as much.
The above dynamic likely has more to do with market structure issues than changing behavior of most fundamental and discretionary investors. But it contributes to a deterioration in liquidity and increase in volatility, giving discretionary investors less confidence to buy the dip, which feeds back into liquidity withdrawal, creating more gaps in the market, etc. At the end of the day ‘buying the dip’ is very similar to ‘liquidity provisioning’ and a decline in one leads to a decline in the other.
Less dip buying would increase the marginal impact of short gamma volatility control funds, including annuities, risk parity funds, etc. As QDS has noted before fundamental investors have largely learned to handle these flows over the last few years, which in part means they’ve learned to buy into the weakness vol control funds create. If those dip buyers are not there, systematic strategies will come to have larger and more sustained impacts.
As a final consideration, it’s worth noting there has been negative correlation in SPX returns day-over-day since 1998 (i.e. mean reversion), but prior to then returns tended to be positively correlated (i.e. gains followed by gains and losses followed by losses - momentum). Coincidently or not, 1998 was also the time when stock-bond correlation turned from positive to negative. While this may be a little conspiratorial, the parallels are notable, and raises the question whether an end to ‘buy-the-dip’ that comes alongside higher rates would help end the structural diversification of stocks and bonds. QDS has written at length about the risks from such a shift (see Is this Starting to Remind you of Something? Oct 4 2018 and Don't Fear A Little Inflation, Yet Feb 26 2018).
None of this is to say that equities can’t bounce sharply on short covering from current oversold conditions – the point is simply that bounces going forward are more likely to be short-lived and should be sold into rather than chased.