pixel

URGENT ALERT: Please beware of fraudulent WhatsApp groups and other groups across Social Media pretending to be affiliated with Anchor and Anchor staff members. Do not engage with these malicious and fraudulent groups in any way. Please direct all queries to invest@anchorcapital.co.za.

Will rising rates kill growth stocks?

The major market narrative at present is that rising interest rates are bad for stocks – growth stocks in particular. The reality is that stocks can go up (and down) with both rising and falling interest rates. In Figure 1, we highlight the Nasdaq (dark blue line, intended to represent tech and growth stocks) and the US 10-year yield (light blue), while the shaded green and red areas are periods of falling and rising interest rates, respectively. Interestingly, you will notice that, over the past decade, the Nasdaq rose irrespective of the direction of interest rates.

Figure 1: The Nasdaq vs US 10-year yields, March 2011 to date

Source: Anchor, Bloomberg

In terms of the argument relating to the absolute level of interest rates, we highlight that the tech bubble took place with the US 10-year yield at around 6%. And the steepest part of the ascent came as yields rose from 4% to 6%. In theory, rising interest rates should put a dampener on valuations, however, in practice, liquidity and animal spirits trump theory.

Figure 2: Will rising rates kill the bull?

Source: Anchor, Bloomberg

That said, the speed (and volatility) of the YTD ascent in yields is important, as it probably would have triggered massively leveraged bondholders to partially unwind their positions, with knock-on effects for other asset classes. The narrative element is also important – to the extent that market participants believe the rates vs stocks and value vs growth narratives, they become a self-fulfilling prophecy, at least temporarily.

So, where do we stand in terms of the growth vs value trade? Factor timing is, almost by definition, impossible. But to the extent that growth (as a factor) was crowded, that extreme in positioning has pretty much unwound – see the blue diamond on the far right of the chart in Figure 3 below. The unwind has been extreme vs history.

Figure 3: JP Morgan momentum crowding indicator – no more signalling risk to momentum

Source: JP Morgan Chase

Growth was also highly correlated with momentum (the latter just picks the strongest performers), but that has unwound (see Figure 4). Now, value is more highly correlated with momentum vs its own history – this ratio has some room before it reaches prior peaks and, potentially, rolls over.

Figure 4: Momentum growth correlation – reversed below the median

Source: JP Morgan Chase

Figure 5: Momentum value correlation on a steady rise

Source: JP Morgan Chase

Not surprisingly, market participants expect value stocks to do well:

Figure 6: A record net percentage of asset allocators think value will outperform growth

Source: BofA Global Fund Manager Survey

Value frequently correlates with cyclicality, which is clearly how investors are currently positioned:

Figure 7: Bank of America (BofA) Global Fund Manager Survey (FMS) shows cyclical consensus is “cyclical” vs “defensive” a year ago –BofA FMS positioning vs historic z-scores

Source: BofA Global Fund Manager Survey

 

On the flipside, investors have cut their tech exposure to levels last seen more than a decade ago.

Figure 8: FMS tech overweight has been cut to its lowest level since January 2009, net % of asset allocators say they are overweight tech

Source: BofA Global Fund Manager Survey

Putting all of the above together, we suspect (no guarantees!) that the worst of the growth sell-off is over. We do not think tech and growth stocks need the 10-year yield to fall, these counters only need it to stop rising at such a rapid pace. As US hedge fund manager David Tepper pointed out recently, the Bank of Japan is not moving its yield targets, which make US treasuries increasingly attractive to Japanese investors. European Central Bank President Christine Lagarde has also re-emphasised her commitment to maintaining a supportive monetary policy regime, further reinforcing the ceiling on the US 10-year.

We believe that the initial yield shock factor is gone, at least for the short to medium term. While not necessarily at a tipping point just yet, positioning risk has shifted from growth to value. This is not to imply that any factor must behave in any particular way prospectively, but markets have a habit of upending consensus narratives and positioning.

 

OUR LATEST NEWS AND RESEARCH

INVESTING IN YOUR NEEDS

Submit your details and we’ll give you a call back to assist and advise you on your investment.

SUBSCRIBE TO OUR NEWSLETTERS

Subscribe to our newsletters to receive regular market commentary, research and updates from the Anchor team. Select between our Individual or Financial Advisor newsletters by selecting the relevant tab below.

WEBINAR | The Navigator – Anchor’s Strategy and Asset Allocation, 2Q24

Anchor CEO and Co-CIO Peter Armitage will host the webinar, provide an introduction to current global and local market conditions and give his thoughts on offshore equities. Together with Head of Fixed Income and Co-CIO Nolan Wapenaar, Pete will also discuss Anchor’s strategy and asset allocation for 2Q24, focusing on global equities and bonds. In addition, Fund Manager Liam Hechter will provide insights into local equities, highlighting some investment ideas; Global Equities Analyst James Bennet will discuss Ferrari and give an update on Tesla, and finally, Analyst Thomas Hendricks will participate in a Q&A with Peter, explaining the 10-year US Treasury to attendees.