Keeping ahead of the Power Curve — Part One

Our goal is to keep you ahead of the power curve in terms of market developments. Almost any analyst can offer a decent set of portfolio recommendations based on what markets will do in the future. The hard part is getting the forecast right.

If you can consistently get the forecast right, then allocations are straightforward. If you can’t do the forecasting properly — either because you have flawed models or are just guessing — then your portfolio allocations may be dangerously out of step, and you’ll be set up for large losses.

Forecasting doesn’t involve a crystal ball. There isn’t one when it comes to markets. It’s not about being smarter than everyone else — there are plenty of IQ points to go around on Wall Street.

The critical skills are knowing where to look for signals of future developments and how to look in terms of predictive analytic models. As we’ve said before, it’s not a matter of predicting the future. It’s more that the future is here today and can be discovered if you dig in the right places with the right tools.

Causation over correlation and randomness

The key to this analytic method is something called path dependence. The idea is that events happening today have a powerful causal effect on events that will happen next month or next year.

The causal effect isn’t guaranteed and isn’t always the same. Exogenous events can intervene to scramble even the best forecasts. But it is powerful and reliable. Applying path dependence is far more effective than using the models that Wall Street and the Fed use, which rely on regressions and correlations.

Path dependence starkly contrasts with the random walk models favoured on Wall Street. The random walk model assumes each event in a series is independent of prior events. If you toss a coin, you could get heads or tails. But each toss is independent of the one before. If you toss the coin 500 times, you’ll get something very close to 250 heads and 250 tails. It’s never exactly that, but it’s always close, and the more tosses you make, the closer you adhere to a 50/50 split.

The coin toss method doesn’t rule out sometimes getting five heads or five tails in a row. Still, the odds against that are high, and the occasional occurrence does not alter the central tendency (or mean reversion) toward a 50/50 outcome. Other examples of random outcomes with predictable odds and mean-reverting behaviour are throws of the dice, card games, and turns on a wheel of fortune.

The random walk is a neat assumption, and it lends itself to all kinds of elegant mathematical models. There’s only one problem — it’s false! Over a century of granular detail on financial and commodity market prices shows beyond doubt that outcomes aren’t random.

The time series of prices shows definite trends, sudden spikes, flash crashes, quick reversals, flatlining, and almost inexplicable gaps that make it impossible to get in or out of a sudden shift in sentiment before it’s too late.

The technical name for this type of pattern is fractal. We don’t have to do a deep dive into the mathematics of fractals. It’s enough to say that such patterns are the opposite of random and bear no relationship to the smooth and steady patterns that randomness predicts.

If the time series of prices is fractal and not random, does this mean we can’t predict them? No. Fractal means irregular and non-random, but it does not mean unpredictable. Again, you just need to understand fractals and use the right tools.

Spreads are signals

The next level of analysis is to consider the implications of different degrees of steepness (also called spread) in different maturity sectors. For example, the Treasury yield curve today shows a rate of 0.5% in one-month bills and a rate of 2.6% in two-year notes. That’s a steep slope between one month and two years. It says the Fed will be aggressive in its rate hikes, something we already know from Fed announcements.

Yet, when we look at the spread between two years and 10 years, we see almost no spread at all. The two-year rate is 2.6%, and the 10-year rate is 2.8% — only a 0.2% difference. This leads us to ask, why is the yield curve steep in the front and flat in the intermediate- to long-term sector? Again, the answer is that expectations of Fed rate hikes are high, but fear of inflation down the road is not high.

One inference of this particular shape is that the Fed will fight the battle of inflation (we know this), but they may win the battle and lose the war — and cause a recession. This kind of warning from the yield curve is more powerful than any analysis from Wall Street or happy talk on financial TV.

We also learn a lot by comparing the shape of the yield curve at different points in time. This is also shown in the yield curve graph above. The blue line is the current Treasury curve, and the black line is the Treasury curve one year ago.

Two signals emerge from the comparison of the two curves. The first is that the curve is higher across the board. One year ago, the curve was stuck at 0% until almost the two-year maturity, whereas today, the curve starts at 0.5% and hits 2.6% at the two-year maturity. Likewise, the five-year maturity was 0.9% one year ago and is 2.8% today. That’s a huge change and reflects both Fed tightening (at the short end) and inflation (at the five-year point).

The other signal of significance is a change in shape at both ends of the curve. The short end of the curve is much steeper today than a year ago. But the long end of the curve is much flatter. As discussed above, steepness at the short end signals Fed tightening, and flatness at the intermediate- to long-end signals no fear of inflation and some fear of recession.

That signal is even richer when we see the year-over-year comparison. The steep-to-flat nature of the current yield curve is a significant change from the 2021 flat-to-steep shape. This signal changed expectations. The 2021 curve might have been interpreted as bullish — low rates would produce higher growth. The 2022 curve is definitely bearish — higher rates will produce recession. This shows how comparing yield curves over time gives us even more information than a snapshot of one yield curve.

Keep an eye out for next Wednesday’s article where Jim delves into yield curves and their impact on interpreting figures in the market.

All the best,

Jim Rickards Signature

Jim Rickards,
Strategist, The Daily Reckoning Australia

This content was originally published by Jim Rickards’ Strategic Intelligence Australia, a financial advisory newsletter designed to help you protect your wealth and potentially profit from unseen world events. Learn more here.

The post Keeping ahead of the Power Curve — Part One appeared first on Daily Reckoning Australia.