Unveiling Macro Mysteries: Why the Fed's Policy is a Gamble

(The following post was originally authored in our Macro section of our Asymmetric Market Update?? delivered on August 3rd, 2024, which is free to subscribe to here)

In our previous macro commentaries, we have focused on the most relevant topics concerning potential market impact, the state of the world, and how to navigate these complicated times.

We discussed (the month before bank fears surfaced and terrified the market) the risk to small and medium banks, given the unequal distribution of excess reserves, even though there were plenty of reserves in the system.

We repeatedly wrote about the mixed economic data. We discussed the “duck economy,” where everything looks smooth on the surface, but there is a lot going on beneath the surface. Beauty is in the eye of the beholder. The headline data is robust, but if you look under the hood, you can weave whatever bullish or bearish narrative you are predisposed to.

We discussed the Magnificent 7 vs. the rest of the stock market. Like the economic data, the stock indices were doing very well; however, upon a deeper dive, the magnificent 7 stocks were doing exceptionally well while the rest of the market was flat to down.

In this edition of Asymmetric Macro, we will tie together all of the concepts we have discussed into a coherent story that starts and ends with monetary policy theory itself.

Monetary Policy

With any data set, you need to define the underlying distribution before you can do any meaningful analysis. We will use three basic distributions for simplicity when describing the world. None of them is perfect, but the point will be clear. Headline economic data is used to describe the overall economy or the economy on average. That makes sense in concept as you cannot custom-tailor economic policy for every individual (to use an extreme example). That would be “unfair” in many lights and impossible to implement in reality. So, we use the aggregate data to describe the state of the economy and, therefore, what monetary policy is best for that aggregate. Let us first go through the three distribution types to describe the underlying population.

Note: We are not writing a PhD thesis here. This discussion is not complete nor bulletproof, as we have limited space. We weave a tale deeply relevant to the current world and the state of economic policy. So, instead of picking apart mundane details, think about the concepts and the potential implications of such at a cerebral level.

Uniform Distribution

As you can see, a uniform distribution is one where every observation (in this case, the socioeconomic position of individuals) is identical. A uniform distribution would be the communist dream. A uniform distribution would also yield the best data set on which to conduct monetary policy. There is no variance if everyone is in the same position, so the “average data” would perfectly represent everyone. Therefore, monetary policy based on this data would be perfect (assuming economic theory is valid and applied precisely according to the rules). We know this is not the case. Communists can dream!

Normal Distribution

The mean, median, and mode are identical in a Normal Distribution. Exactly half of the observations (in this case, the socioeconomic position of individuals) are to the right of the center, and half of the values are to the left of the center. This distribution would imply a socioeconomic density highest around the mean with declining amounts of privileged or underprivileged individuals as the deviation from the mean increases. With a dominant middle class and reasonable wealth distribution (as the US had much more balance in the not-so-distant past than now), “average data” is also functional. Though it is not perfect, with density still centered around the mean, monetary policy conducted on this data is reasonable as it captures the majority of the population’s situation (though monetary policy would not be relevant for the tails of the population; that is a relatively small percentage in a normal distribution.

Bimodal Distribution

A bimodal distribution has two modes. In other words, the outcome of two processes with different distributions is shown together in one set of data.

This characteristic of a bimodal-ish distribution has been popping up in all aspects of our world recently. Let’s look at a few relevant examples that we have previously written about.

The Unequal Distribution of Excess Bank Reserves

In Asymmetric’s February 2023 release, we wrote, “The answer seems to be that though there are plentiful excess reserves in the system, they are not evenly distributed. The reserves are concentrated in the money center banks (JPM, etc.)”.

So, though the headline amount of excess reserves was beyond plentiful, we had a banking tail event that caused the Federal Reserve to institute emergency facilities to fund many banks that lacked sufficient reserves. Several major banks collapsed before the dam was plugged with this facility. Why did this take everyone by surprise? The excess reserves data was topline data and did not consider the underlying distribution of where those excess reserves were held. Many had none. Some had the majority. That is a bimodal distribution. The aggregate data alone did not tell an accurate story of the banking sector. So the distribution here was crucial, yet ignored.

The unequal distribution of reserves and subsequent emergency funding facility manifested into the weak paying considerable interest expense to fund their balance sheets and to maintain/gain deposits. The strong (JPM) received significant interest income on their excess reserves. Robbing the poor to pay the rich. One could argue this is the penalty for poor management, and one would not be wrong. But it still leaves you with a bimodal distribution going forward. This situation continues to become even more bimodal, given the dynamics.

Small Business vs Megacaps

In Asymmetric’s July 2024 update, we posted the following graph:

Magnificent 7 vs the other 493 Companies, S&P 500 and the Russell 2000

Looking at the Magnificent 7 versus the rest of the equity market (particularly the Russell) shows something of a bimodal distribution as well. You have a very healthy, outperforming group of large companies; then you have the smaller companies that enjoyed nowhere near the success of these megacaps.

One could argue this is the capitalist outcome of creative corporate destruction, and one would not be wrong (we will ignore the implications of monopoly/oligopoly industries for this discussion). Regardless, given the dynamics, it still leaves you with a bimodal distribution going forward that continues to become even more bimodal (or a series of monopolies in the boundary condition).

Some of these outcomes are attributable to the power of scalability in technology. Once you dominate an area, you suck the business potential and capital out of your competitors. Because of this, the large companies ended up with large cash piles from record profits. They bought back stock and received significant interest income on their cash piles. The smaller companies ended up with more onerous debt piles (not cash-rich), paying large sums of interest to stay alive. Robbing the poor to pay the rich.

Socioeconomic Distribution

We chose the graph below as a convenient illustration of a bimodal-ish distribution in socioeconomic status. The data set has two different modes representing the fragmentation in society. Is looking at the average credit score here useful? Not at all. That’s the point. We are conditioned to look at average data, but with bimodal distributions, that can be, at a minimum, not useful and, at a maximum, very deleterious and misleading for analysis.

We could add texture around the distribution of individual savings, debt/credit service expenses, etc., but we all know what it would show: a bimodal distribution. And just like the examples above, those paying high-interest expenses are in big trouble. Those with excess savings are reaping the benefits of these same high interest rates. Robbing the poor to give to the rich.

The well-to-do are doing fine, per the above graph.

Those with less disposable income are not doing so well.

Tying It All Together

What do the above three examples have in common? Paying vs. receiving interest gives two diametrically opposed outcomes—the poor get poorer, and the rich get richer. It is that simple (not really that simple, but kind of). Wealth/assets are being siphoned from the weak to the strong.

Why is all of this relevant? Monetary policy is based on aggregate data. ON AVERAGE, everything has been good and still seems ok. That said, one mode of this distribution is in severe pain. High rates advantage the other mode of this distribution. So, by keeping rates high and waiting for the average data to wane, they are simply crushing the weak more than they are advantageous to the strong. It is twisted when thought of in that context.

Why does the wealth divide keep getting wider? Because the way monetary policy is administered exacerbates and increases the wealth divide. This is not a white paper on the virtues of wealth redistribution for the greater good. Still, it seems that across most major areas of our economic lives, the wealth divide will widen until we end up with a collapse of some sort, a debt jubilee, or some other tail-type event to be determined.

In Conclusion…

In our view, the Fed should have cut in July.

Employment has peaked and rolled clearly.

Inflation is 2.5%, is declining quickly, and is expected to be at the 2% target by year-end.

Yet real rates are currently at 3%. Historically, they run at approximately 1% in a steady state, healthy economy.

So what is the Fed doing?

They are looking at the aggregate data without regard for the underlying distribution.

This is how a policy error happens.

The wealthy/cash-rich enjoy higher rates for interest income (not to mention assets are near all-time highs). The cash poor are getting devastated with interest expenses. Given this insensitivity, or even benefit, of the one mode (cash-rich individuals, corporates, etc.) to higher rates, the Fed is implicitly waiting for the lower socioeconomic tiers to deteriorate further to bring the average data down to targets. Sorry poors, you wear the pain with the least upside.

If the Fed allows “tight monetary policy” to continue (their words), they run the risk of severe employment issues and the hollowing out of small businesses. Once that is broken, history says it is tough to reverse. They run the risk of a hard landing.

Everything is fine until it really isn’t. Very slowly, then all at once.