Category: Contrarian Corner

What Does The Manufacturing PMI Tell Us About Forward Stock Returns?

Reading Time: 2 minutes

When you think about all the articles being written about shortages and fears of inflation, it seems like the US economy is doing very well. You can’t really have those things without consumer demand. Indeed, some are calling for GDP growth of 8% in 2021 and a big increase in inflation.

I’m a little cautious on the GDP growth in 2021 causing sustained inflation (mainly because its high growth lapping a year that was beaten down) (as I previously wrote about). Let’s not confuse a one-time increase in prices with inflation…

But I also get pretty cautious when everything I read is all the same – “a boom is here.”

One piece of data that gets thrown around is ISM Manufacturing PMIs. The latest reading was 64.7%! It’s well above pre-COVID levels.

As they say on their website:

A Manufacturing PMI® above 43.1 percent, over a period of time, generally indicates an expansion of the overall economy. Therefore, the March Manufacturing PMI® indicates the overall economy grew in March for the 10th consecutive month following contraction in April 2020. “The past relationship between the Manufacturing PMI® and the overall economy indicates that the Manufacturing PMI® for March (64.7 percent) corresponds to a 6.2-percent increase in real gross domestic product (GDP) on an annualized basis,” says Fiore.

Now, a PMI is a “purchasers managers index” and is basically a survey from a wide array of companies in manufacturing in the US. It basically is asking, “are you growing or shrinking?” across a wide array of topics.

Investopedia says, “when the index is rising, investors anticipate a bullish stock market in reaction to higher corporate profits.”

Actually – it can really be used as a CONTRA indicator.

I went back through the data to 1960 and checked the 6-month, 12 month and 18-month S&P500 returns after the manufacturing PMI was at certain levels. It’s not perfect, but its something.

I’ll just put the data out there – would you rather swing hard with the index around 65? Or below 43? There aren’t many cases when it goes above 70, but that generally does not have a good track record.

Look, I try not to market time. I’m typically nearly-fully invested. But I also understand there’s a time to own some names (e.g. cyclicals) and time to eh… maybe cool it and wait for a better day.

3D Printed Homes Aren’t New

Reading Time: 3 minutes

I’ve seen many articles recently about the future of 3D printed homes. It sounds extremely exciting – pick a design, and the 3D printer goes off and creates the home with less waste and less labor and hopefully less price. There have been several recent articles such as this home listed on Zillow for a low price, this new startup that’s starting to build 3D printed communities in the desert,  and these 3D printed homes in Austin which aim to be a more mid-tier offering.

3d printed homes

Wait a second – aren’t 3D printed homes this just another name for manufactured housing? Seriously. Compare the picture above (3D printed home) with another graphic of a manufactured home from Cavco.

Cavco manufactured home looks just as good as 3D printed homes

Tying this into this blog, which is investment related, I just did a post on Skyline Champion and Cavco, which prefab homes in a manufacturing site and then send it along to the plot of land desired by the customer. The difference seems to be having a cool name for your process and doing most of the manufacturing onsite.

This home listed in Long Island was for sale for $300k – lauded for its ability to sell at a material discount to homes in the area. But my guess is that’s probably the same price / more expensive than what manufactured housing offers from Clayton, Cavco or Skyline. And the 3D printed house looks…. mehhhh. In this case, the true cost of this home likely came down to lot value in Long Island.

Let’s call a spade a spade…

I struggle to see how 3D printed homes will disrupt current manufacturing processes, especially in comparable product categories. A lot of these 3D printed homes still need fabrication work onsite after they are completed, too.

Here is the cement-based 3D printed home. The machine builds the structure by adding layer and layer of cement, but it looks like the guts of the home are still fabricated without machines. Not to mention you still need someone to install windows, doors, cabinets, countertops, and electrical work.

My guess is that, today, you can only build these structures where the land is very flat – otherwise it will throw off the machine. Technology improves, but it will improve in the manufacturing process in “traditional” areas too.

Many prefab techniques already exist and are continuing for onsite construction. For example, distributors like BuildersFirstsource often assemble the trusses for a home, or they will precut the wood ahead of time so the frame can easily be stood up on the home (BuildersFirstsource has a brand called Ready Frame which is an interesting watch if you have the time).

This same thing happened in Japan, where manufactured housing and 3D printed homes are more commonplace. Much has been written about applying what Japan does to the US.

But outside of low-cost housing, however, Americans desire too much choice and they might as well choose a Clayton Home over a “Google-backed startup” 3D printed home. Japan differs a bit too, in that their homes depreciate over time, whereas everywhere else they appreciate. So it makes sense to build cheap and quickly, raze it later, and start fresh.

We clearly need housing solutions. There is a housing shortage in the US, as I discussed in this post. I worry about that shortage leaving groups of people in the US behind. And I am afraid our employment trends mean that the shortage in construction trades will likely get worse. This will continue to drive up the cost of building. It’s no wonder why there are growing calls to democratize housing.


But solutions have been discussed in the US literally for 90 years. Architect Buckminster Fuller had the idea for a “Dymaxion House” which was a aluminum, grain silo-looking house that could be shipped and easily assembled with less waste. Frank Lloyd Wright did as well. The idea being that if the automobile was being democratized, so should housing.

Obviously, it never panned out.

The undertones sound exactly the same as today as the 1930’s, though. We need to improve cost, we need to save energy, we need to become more efficient. Democratize housing.

We have solutions to that today, but either consumers aren’t choosing it, there are zoning issues, or something else.

I personally have trouble seeing 3D printed homes offering a meaningful solution in the near term.

Contrarian Corner: Inflation is a Consensus Bet. Look at Deflation $TLT

Reading Time: 5 minutes

I’m thinking of starting a new segment called, “Contrarian Corner.” In these posts, I will try to point out the other side of a company perception, trade, or view that I see as pervasive in the market.

When everyone crowds to one side of the boat, there are typically better opportunities to sit on the other side.

Right now, it seems “inflation is coming” is a pretty consistent view. (Somehow, the people shouting inflation think they are contrarian?) Every report, article, or tweet is talking about it…. I really try to avoid macro talk, but this is too good to pass up.

My favorite example of this is pointing to lumber prices as an indicator of runaway inflation.  This totally ignores prior supply / demand dynamics that led to this surge. To me, its picking a data point to support a view.

I follow lumber prices, so I know why they are up. Canadian lumber is high-cost supply. The crash in prices in 2018 meant many mills to the north were unprofitable and were curtailed. Add in forest fires and impact from the Pine Beetle and supply was constrained. Lastly, Canada implemented caribou protection which curtailed logging activity. Lumber prices were still very low so even mills in the US shut. Now that housing has come back strong, this caught supply off guard and prices surged.

Is that inflation? Or is that a short-term supply demand imbalance? Prices are now at a level where everyone can make money if they can get supply back online. Would you make a bet with me that lumber prices will be higher than where they are right now in 3 years?

There are other examples of this. I’ve discussed oil (prices are up and rig count is at multi-year lows) as well as housing (underinvested post-GFC), and used car prices. Each of these are specific  supply / demand issues or the bullwhip effect.

Is that persistent inflation? Is a one time rise in prices due to a demand shock (COVID lockdown), that resulted in a supply shock, inflation?

The reason why *persistent* inflation matters is because that is what is going to move long-term interest rates.

So Why Deflation vs. Inflation?

First, its that the government’s use of debt to stimulate is suffering from declining marginal returns. Second, its that M2 is NOT what drives inflation. Its also velocity, which continues to decline.

As the National Bureau of Economic Research stated in a 2010 study (my emphasis added):

The median growth of the 20 advanced nations in this study fell by half as their debt levels moved from less than 30 percent of GDP to 90 percent or more. The drop-off was particularly significant at the 90 percent threshold: between 60 and 90 percent of GDP, median growth was still 2.8 percent; above 90 percent it was 1.9 percent. The drop in average growth between countries with debt ratios of 60-90 percent of GDP, and those above 90 percent of GDP, was even greater: 3.4 percent to 1.7 percent

What happens when you go from 100% to 120%? Japan is approaching 200% debt to GDP and we all know the impacts there (their central bank also straight up buys equity ETFs)

Essentially, the marginal benefit we  get from adding a new dollar of debt is going down. And has been for quite some time.

Yes, the coronavirus stimulus was big. But a lot of it also went to plug a big hole in the economy.

We had stimulus checks. That put money directly in the pocket of consumers, but it didn’t create a new income stream for them. Wages didn’t go up and in my view and so the spending will be a 1x boost in some select sectors. Unless all the debt we just used goes to create a new income streams, all we’re doing is exchanging current consumption for future consumption.

Velocity of money is going down.

I’m going to let Dr. Lacy Hunt explain the next bit. For context, he’s a manager on Hoisington bond fund and has been right on bonds for about 40 years (i.e. he’s been long duration). I highly recommend his investor letters. All of these are quotes from his Q1’2020 letter, with my emphasis added:

  • When the Fed buys government or agency securities from the banks, holdings of government debt declines and the banks’ holdings of deposits or reserves at the Fed go up.
  • The bank balance sheet is unchanged except that the banks are selling government paper of longer maturity and they receive an overnight asset at the Fed. Those deposits do not circulate freely within the economy. (Diligent Dollar Note: QE is not just printing money)
  • If the Fed’s purchase of the debt is from non-bank entities, there will be a transitory rise in M2. Further M2 expansion from that new level will depend on the banking industry. The banks high level of reserves at the Fed will result in no further increase in money unless they and their customers make the collective decision for new bank loans to be originated and the loans are used to expand economic output
  • This is what happened in 2010-11. M2 surged transitorily to a nearly 12% rate of growth along with an increase in loans. The money and loans were used to shore up financial conditions rather than channeled into the purchase of new goods and services. As such, the velocity of money fell dramatically, and the Fed’s purchases of securities did not lead to increased economic growth and inflation. After financial conditions were stabilized, the depository institutions held large amounts of excess reserves.

I feel like the first two bullets need re-emphasizing, because a lot of people associate QE with money printing. Joe Weisenthal put it (again, my emphasis added):

When the Fed buys a Treasury, what it’s really doing is replacing one kind of government liability (maybe a 10-year Treasury) with another kind of government liability (an overnight reserve held at the Fed). If you’re a bank that sold a Treasury to the Fed, you’ve given a long-term asset that yields something for a short-term asset that yields something else. No new money has entered the system. The government doesn’t have any less debt. All that’s happened is that the consolidated government balance sheet (the Treasury and Fed combined) has shortened the term structure of its liabilities. After the Fed buys a Treasury there’s less long-term debt outstanding and more short-term debt outstanding. That’s it.

Bottom line: in order to actually get a boost to GDP and inflation, you need the velocity of money to go up.

All these headlines of GDP growth in 2021 tend to be missing the point. If we have 8% growth in GDP following a year where GDP was down 3.5%, then you just have 2% growth on a 2-year basis. That’s not that great considering all the debt taken on. And then you actually need the banks to lend out the capital, but that depends on risk they see in the market, returns, and whether people need the capital for investment.

Add in worsening demographics (and I think its possible we are sitting where Japan and Europe are sitting 5 years from now.

I think what will happen is inflation expectations will continue to rise short-term as the economy re-opens and we have some supply constraints that gives further pseudo-inflation scares, but long-term the writing will be on the wall.

So bottom line: Do I think reflation will happen? Yes. We will be lapping a serious decline in our economy and there are supply constraints. Do I think those supply constraints will be overcome? Yes. And therefore, I think we will continue on the longer-term deflation trend.