Hedgeye Founder and CEO Keith McCullough joins Alan Dunne on the Top Traders Unplugged podcast for an in-depth look into Keith’s distinctive approach to global macro analysis and tactical trading.
Below is an excerpt transcribed from the conversation. Listen to the entire 62-minutes here.
The conversation explores Keith’s background as a hedge fund manager and his views on how markets have evolved, especially in relation to option flows. Keith breaks down his four-quadrant framework for macro analysis, explaining how pinpointing the economy’s position within these quadrants is as much an art as it is a science.
His framework currently signals continued strength in the US dollar and US equities. A fundamental aspect of Keith’s approach is his respect for market signals and the need for these signals to confirm his macro views. He also shares valuable insights into the behavioral challenges of investing and emphasizes the importance of a disciplined, repeatable process to drive long-term success.
Alan Dunne: And in the Master the Market book you mentioned, you suggest that markets are more macro driven now. Is it that you’re saying that stocks are more influenced by macro factors and fundamentals, is that it?
(Download Keith's free "Master the Market" eBook here.)
Keith McCullough: Definitely. So, we call them the Quads. So again, I invented this thing called the Quads. The basic assumption is that the economic cycle is being driven by reported data, and "The Machine" is going to adapt to whatever those macro conditions are. I have a two-by-two model: the rate of change of growth, the rate of change of inflation.
The biggest risk in markets is when we hit what we call Quad 4. We have four different quadrants. When the market is sniffing out a rate of change slowdown in both growth and inflation, whether or not you’re going to have a recession is irrelevant. What ends up happening is that the dollar goes up, bond yields go down, utilities go up, rate-sensitive equities go up.
The sector styles obviously take note, and you have your macro positioning staring you right in the screen. So, when we go through this, the most important things that happen, from a macro perspective, are perpetuating the incremental flow of "The Machine." It either stays with the trend of whatever Quad you’re in or the Quad shifts. And that’s where the big moves happen.
We just had a huge one in inflation expectations. Our model picked that up. Instead of Quad 4, the opposite of that is Quad 2. In Quad 2, you have the rate of the rate of change of growth and inflation accelerating at the same time. Bond yields go up, the dollar goes roof, and you get this massive breakout in small caps. Financials love Quad 2.
So, we backtested, basically, everything that ticks in macro against these four Quadrants. We have a playbook. We don’t always go with every single thing that backtests. It has to have a signal that supports it. And that’s really my principal orientation. Where’s the signal, where’s the quad? And then I go from there.
Dunne: Okay, great. So just to recap, it’s kind of four Quads. One is kind of Goldilocks growth up, inflation down. Two is reflation, so higher growth, higher inflation. Three is stagflation, so growth weakening, inflation staying elevated. And four is deflation, so growth deteriorating and inflation deteriorating.
McCullough: Exactly. The devil is in the details, as you said. I’m an Irish Catholic, I believe in the devil. He’s there—it’s in my machine every day. So, you’ve got to really pay close attention.
Dunne: What are the pushbacks in all of this macro-driven analysis, I guess, particularly if you’re modeling it? Is it the challenge of, okay, it doesn’t happen too often, but when something like, say, Covid comes along and the macro data is slow to pick up what’s evolving in the economy, was that problematic for your models?
McCullough: No, it would be if I didn’t have my signaling process, that’s for damn sure. The signal front-runs the quads. To your point, economic data is reported on a lag to when it’s actually happening. Sometimes it’s a month, sometimes it’s a week, most of the time it can be a quarter.
The signal is constantly telling you what its probability-weighted view is of the future economic reality that’s about to be reported. The pandemic is an excellent example. I didn’t know anything about pandemics, but my signal in January of that year went explicitly Quad 4. And this was well ahead. So, I’m positioned for a pandemic, and I didn’t even know a pandemic was going to happen.
Markets discount the future, and they’re very good—particularly one to three months ahead of time—at giving you plenty of opportunities to learn about where you’re about to lose money. I always say that if I’m losing money in the Quad that was working, that means the Quad’s changing. So, I better change.
Dunne: Yeah, and you mentioned the signal lead. When you’re talking about the signal, is that this idea of Trade-Trend-Tail?
McCullough: Yeah, well I mean it’s my own so it’s not Dow theory, it’s not whatever. I call it multi-factor, multi-duration. Multi-duration, you just nailed it.
I use three different windows of time that I care about. What is the immediate term trade, which is three weeks or less? I built it that way, by the way, because I wanted to front-run my boss. Trends are three months or more. And tails, I call them tail risks, are three years or less. We’re far less accurate on tails because it’s further out in time.
On the multi-factor point, which is the real differentiator, I use the relationship and I built it as a fractal model that uses three factors. When I say multi-factor, multi-duration, the three factors are price, volume, and volatility, the rates of change they’re embedded within, and their relationships and how they’re tied together.
*****
