Summary: Asset-First Economics tells us to focus on the Balance Sheet, not GDP. Interest rates are determined by tangible asset inflation, not the CPI or the PCE. Existing home price inflation of 1.9% means the Fed should not raise rates. Worries about the budget deficit and interest rates are misplaced. There is no doubt that AI will raise productivity and GDP and that AI/Data Center Capex is the main driver of both output and earnings today. But we should remember that information businesses have increasing returns and declining costs, which makes them winner-takes-all natural monopolies. Natural monopolies only have one winner. The first guy that gets to scale will drive prices to zero and kill all of its competitors; it will earn revenues not by selling product but by monetizing the eyeballs of its massive user base. Regardless of which company wins the race, its founder will be anointed as a true genius. Its competitors growth plans will vaporize, along with their capital spending plans. The ultimate script for the impact on stock and bond prices when markets figure that out is still to be written. It will be a lot more volatile that the one you read in the newspaper.
Here are a few comments on some things I am actively working innow and will write about in future posts:
Inflation. Both CPI and PCE have serious measurement errors. Housing costs for existing homeowners are dramatically overstated by adding a made-up (imaginary) measure called “Owner’s Equivalent Rent.” It should be removed from both indexes. Correcting that mistake would reduce the June Core PCE inflation rate from 2.6% to 2.2%, and July Core PCE inflation from 2.5% to 2.2%. More on that in my next post.
Fed Policy. I like what Kevin Warsh is doing to get rid of the Dotplot and reduce the endless string of Fed Governor lunch talks we call Fedspeak. If I were advising him what to do at the next FOMC meeting I would say to leave rates right where they are.
And I think shrinking the Fed’s balance sheet, getting the Fed out of the business of financing budget deficits, and returning to the days when the size of bank reserves was an important tool for the Fed, are good ideas.
AI. It is beyond question that the adoption of AI will raises productivity and drive growth higher. Like miles per hour, GDP’s dollars per year is just a measure of speed. AI makes work get done faster. GDP has to go up .
The construction economy is booming, due to the data center buildup, which is both helping employment and putting upward pressure on inflation through its impact on construction costs and information equipment prices.
Ultimately, AI will drive inflation lower for a given level of interest rates.
The impact of AI on jobs depends on how fast AI adoption takes place. Over long periods, improvements in technology have important impacts on where people work, but not whether they have jobs. (In Ben Franklin’s time, 93% of Americans were farmers, When I was born, 50% were farmers. Today just 3% are farmers. The former farmers are just doing different work.) The only real issue is how fast people need to, and are able to, change. Here are a few things to keep in mind:
The booming information sector is having a huge impact on corporate earnings and stock prices. Some of that is fake—the reported profits announced by companies that come from holding shares of each other’s stock. But most of it is real.
Information companies have fat gross margins by their very nature. They are the declining marginal cost, increasing return, businesses that Brian Arthur writes about at the Santa Fe Institute. That makes them natural monopolies, which increases the concentration of wealth, turns the founders of the winning companies into imaginary geniuses, and makes a lot of people mad.
It also pushes profit higher as a percent of GDP. (Earnings growth estimates have increased by 25% this year.)
It pushes labor’s share of income lower, increasing people’s sensitivity to food and gasoline, and health care prices. And it increases the percentage of consumer spending that comes from the owners of all that stock, making GDP more volatile. This is the source of the political breakdown that has plagued us for the last decade.
And it has created meaningful Bubble Risk that people are beginning to talk about. One day investors are going to figure out that industries that are natural monopolies only have one winner. That means the other companies that are not winners won’t be able to follow through on the things they have announced to shareholders and will have to curtail their plans, i.e., their Capex plans are going to shrink. That’s when we should worry about the reaction from bond and stock markets. More to come on this.
My work on economics and investing comes from an analytical framework that I call Asset-first Economics. It rests on a simple idea: there are two economies, not one. The first economy is our $618 trillion balance sheet economy that measures the value of what we own, what we owe, and our net worth. The second is our $32 trillion paycheck economy that measures our income and the value of the work we do over the course of a year. Economists all write about the paycheck economy. For investors, the balance sheet economy is the only one that matters.
All of the major economic and financial storm systems that matter for investors originate in the balance sheet when changes in inflation, tax rates, tariffs, regulations, and global conflict trigger the massive changes in demand for tangible and financial assets that drive the tsunamis we experience in stock, bond, real estate, and commodity prices.
I have written extensively about two ideas. One is that inflation impacts interest rates by triggering changes in investor demand for tangible and financial assets, not via flows of funds like savings, investment, and budget deficits. As such, the right inflation measure is tangible asset inflation, which measures the capital gains yield on the $191 trillion stock of real assets that we call net worth, not the CPI or the PCE. A pretty good proxy for tangible asset inflation is existing home prices because homes make up the lion’s share of household net worth. The most recent numbers for July show that existing single-family home prices increased by 1.9% over the past year, slightly below the Fed’s 2% inflation target, which is why I would advise the Fed not to increase interest rates at their next meeting.
The second idea concerns how to think about the impact of budget deficits on interest rates. In Asset-First Economics, comparisons of deficits, government debt, and GDP mean nothing. What matters is whether investors want to hold the debt. That’s largely a matter of investor net worth and of the relative returns of different types of assets.
Today’s $34 trillion national debt makes up only 5.5% of America’s immense ($618 trillion) stock of total assets and 18% of our ($191 trillion) household net worth. All of those Treasury securities are willingly held (demanded) by investors at today’s interest rates. And the demand to hold Treasury securities is increasing every year in line with our growing net worth, which has grown by an average 7% per year over the past 50 years. That means next year investors will want to own 7% (+ $2.5 trillion) more Treasury securities than they already own today. In other words, the first $2.5 trillion of new Treasuries issued next year—next year’s budget deficit—are already spoken for. The budget deficit would have to be bigger than $2.5 trillion to put any upward pressure on interest rates at all.
The big numbers in the preceding paragraph come out of the Fed’s quarterly Z.1 Report, the single most valuable economics report published today. You should know, however, that the report dramatically understates both total assets and net worth for two reasons. First, the asset figures in the Z.1 balance sheet exclude the value of all non reproducible tangible assets, including the roughly 700 million acres of land, the energy and mineral rights attached to it, and the Continental Shelf, owned by Federal, State, and Local governments. And second, because the Fed’s Net Worth calculation counts the market value of outstanding shares of stock as Liabilities of the company’s that issued them, i.e., they show the corporate sector as having a massive negative net worth. But corporate equities are not debt; they are claims on the future free cash flow of the companies the shareholders own. Setting these two accounting matters right would increase the resulting net worth figures by at least half.
Before you start calling me a Commie Socialist Lefty, though, let’s get one thing straight. None of this means I’m in love with government spending, deficits, or the national debt. It doesn’t mean the budget deficit or national debt aren’t higher than I’d like them to be—they are. It doesn’t mean that spending, the budget deficit, and the national debt won’t go vertical when you add in the off-budget promises like future social security and Medicare claims that the politicians conveniently forgot to include in the budget. And it doesn’t mean there won’t be huge tax increases and/or benefit cuts when those expenditures show up down the road—there will, although they will likely be in the form of wealth taxes, not income taxes because there won’t be much income left to tax. It just means that we should stop whining about how the deficit is going to explode interest rates and focus on getting our accounting right and whether the stuff the government is spending money on is worth its cost.
In my next posts I will give you some of the technical details of the ideas above. I’ll show you how the BLS chose exactly which three components in the PCE index to reinvent. I’ll explain why we should change the way we calculate the CPI. And I’ll show you a graphical analysis of the logic behind my budget deficit, interest rate statements above.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.