A single company is about to go public at a valuation between $1.75 trillion and $2 trillion. One company. Not a country. Not a continent. A rocket-and-satellite-and-AI business that started in a Los Angeles warehouse in 2002 with one founder and $100 million of PayPal money.

That company, SpaceX, filed its IPO prospectus with the US Securities and Exchange Commission on May 20, 2026. It plans to begin trading on the Nasdaq under the ticker SPCX as soon as June 12. The man who owns most of it, Elon Musk, holds around half the equity outright and controls 85.1% of the votes through a class of super-voting shares. Inside the company's own filing, his future stock awards vest in tranches all the way up to a $7.5 trillion valuation milestone. Polymarket and Kalshi, the prediction markets, both put his odds of becoming the world's first trillionaire before 2027 at over 70%.

Now hold that next to a number you'll still hear quoted on every business news segment, every week, as if it tells you something useful about the country. "GDP grew 2.4% last quarter." "The economy is strong." "The recovery continues."

Those numbers and that company are not telling the same story. They aren't even speaking the same language. And the longer we keep pretending they are, the more useful the lie becomes to the people it benefits.

What follows is the detail. What GDP actually is, what it measures, where it came from, why it became the headline number for everything, and why it has now stopped describing the world it was built for. Once you see what GDP measures and what it hides, the headline reads differently. Different questions surface, about the economy you and I and everyone we know actually live in.

Stop using GDP as if it tells you something about people. It doesn't, and at this scale it actively hides what's happening. When someone quotes it to you, hear what they are leaving out, every single time.

This is going to take a minute. Stay with me.

What GDP Actually Measures

Gross Domestic Product is the total dollar value of everything produced inside a country in a year. Goods, services, the lot. It was developed in the 1930s by an economist named Simon Kuznets at the request of the US Congress, in the wake of the Great Depression. Congress wanted a single number that would tell them how big the economy was, how fast it was growing or shrinking, and whether their policies were working. Kuznets gave them one.

He also warned them, in the same report, that it was a terrible measure of how a country was actually doing. He wrote, almost in the same breath as he handed them the number, that "the welfare of a nation can scarcely be inferred from a measurement of national income."

He spent the next forty years of his career trying to get policymakers and the public to stop using it the way they were using it. They didn't listen. They had a single number now, and a single number is what political coverage and economic punditry love above all other things.

The number does exactly one thing, and it does it pretty well. It tells you the size of the pie. It says absolutely nothing about who eats it.

A country can have a soaring GDP and a population sliding backwards at the same time, and the headline number won't blink. A country can have all of its growth landing in the bank accounts of a few hundred families and GDP will report that as a strong year. A country can hollow out its middle class, export its manufacturing, watch its workers' real wages stagnate for forty years, and GDP will go up the entire time, because the dollar value of the things produced is still growing, regardless of who ends up holding those dollars.

In 2009, the French government commissioned a high-level commission to look at this exact problem. It was chaired by Joseph Stiglitz, a Nobel laureate in economics. The other members included Amartya Sen, also a Nobel laureate, and Jean-Paul Fitoussi. They produced a report called Mismeasuring Our Lives: Why GDP Doesn't Add Up. The title is doing a lot of the work. Their conclusion, stated as plainly as economists ever state anything, was that GDP is "not wrong as such" but is "wrongly used," and that countries that continue to use it as a proxy for societal wellbeing will keep making policy decisions that look good on the headline number and feel disastrous on the ground.

That report came out seventeen years ago. The world's economists know about the GDP problem. The world's politicians and business reporters keep quoting it anyway. There is a reason for that, and it isn't an oversight.

The Split That Started In 1979

For thirty years after World War Two, something remarkable happened in the American economy. When the productivity of the typical American worker went up, the pay of that same worker went up roughly in step. The pie grew, and the slice grew with it.

This was so consistent, so reliable, that economists at the time treated it as a kind of natural law of capitalism. As workers got better at producing things, they got more money for producing them.

Then, starting in 1979, those two lines came apart. And they never went back.

The Economic Policy Institute, a Washington think tank that has tracked this for decades, publishes the numbers every year, and they don't move much because they don't have to. Between 1979 and 2019, productivity in the United States grew 59.7%. The pay of the median American worker, over the same forty years, grew 15.8%. Productivity grew 3.5 times faster than pay. If median compensation had simply kept pace with productivity over that period, EPI calculates, the typical worker would be making roughly $9 more per hour today than they actually do.

That gap, between what you produce and what you get paid, didn't open by accident. EPI is unsparing about this. They attribute the divergence to a specific era of policy choices, most of them clustering between 1979 and the early 1990s. The federal minimum wage was allowed to erode against inflation. Antitrust enforcement weakened. Unions were systematically undercut. Industries were deregulated. Globalization was pursued on terms set by capital, not labor. Tax rates at the top were slashed. Macroeconomic policy began to tolerate higher unemployment to keep inflation low. Wage theft enforcement collapsed.

You will sometimes hear this called the Reagan era. That's a defensible shorthand for the broader policy direction, though the inflection point itself predates Reagan's January 1981 inauguration by about eighteen months. The 1979 break is real and well-documented in the data. The decade that followed accelerated it. Bill Clinton's administration didn't reverse it. George W. Bush extended it. Obama largely held the line. Trump, in his first term, accelerated the part that benefits capital. Whoever happens to be in office, the trajectory has held for forty-five years.

So when GDP grew during that period, and it grew significantly, where did the gains actually land?

They landed in the hands of the people who own assets, particularly equity in publicly traded companies. They did not land in wages.

CEO Pay Tells The Same Story In One Number

If you want the wage divergence collapsed into a single comparison, this is the cleanest one I know. Between 1978 and 2023, realized CEO compensation at the top 350 US firms grew 1,085%. Median worker compensation, over the same period, grew 24%.

Read those numbers again. CEO pay grew more than a thousand percent. Worker pay grew a quarter of one percent for every year. Forty-five years of work.

The CEO-to-worker pay ratio in 1965 was 21 to 1. By 2023, the ratio at S&P 500 companies was 290 to 1, according to EPI. The AFL-CIO's 2024 Executive Paywatch report put the ratio at 268 to 1 using a slightly different methodology, and their 2025 report updated it to 285 to 1 for S&P 500 companies in 2024. The median S&P 500 employee, by AFL-CIO's calculation, would have to start working in the year 1740 to earn what the average S&P 500 CEO took home in a single year of 2024.

At some companies it is more grotesque than that. The Institute for Policy Studies, in its 2025 Executive Excess report, tracked the 100 lowest-paying companies in the S&P 500 specifically. The average CEO-to-worker ratio at those firms was 632 to 1 in 2024, up from 560 to 1 just five years earlier. Starbucks topped the list at 6,666 to 1. The Starbucks CEO took home $95.8 million in 2024. The median Starbucks employee earned $14,674. That isn't a number from a developing country. That's the United States. That's the company you got your coffee from this morning.

Tesla's board has proposed a pay package for Elon Musk valued at up to $1 trillion over the next ten years. One trillion dollars. To one person. For running a company.

CEO pay is, in itself, a relatively small slice of total national wealth. You could confiscate every dollar of S&P 500 executive compensation and it wouldn't close the bottom 50% gap. The reason it matters isn't the dollar volume. It's what the ratio reveals about who the entire system is now built to reward. The CEO pay number is the canary. Whatever it is doing, the broader divergence is doing more of.

One Company, Twenty-Eight Trillion Dollars

Now back to SpaceX, because the IPO sharpens all of this to a single point.

In its S-1 filing on May 20, the company stated, in its own words, that it has "identified the largest actionable total addressable market in human history." It then put a number on that market. $28.5 trillion. The breakdown, also in the filing: $370 billion in space-enabled solutions. $1.6 trillion in connectivity, anchored by Starlink. $2.4 trillion in AI data centers in orbit. $760 billion in future user subscriptions. $600 billion in digital advertising. $22.7 trillion in what the company calls "enterprise applications."

The total US gross domestic product, for all of 2024, was approximately $29 trillion. One company, in its IPO prospectus, has claimed an addressable market roughly the size of the entire American economy. Not as a long-shot aspiration. As the basis on which it is asking public investors to buy its shares.

Hold the IPO target ($1.75 to $2 trillion) next to a few real countries' annual economic output. Spain produces roughly $1.7 trillion of goods and services a year. Mexico produces about $1.85 trillion. The Netherlands produces around $1.2 trillion. Saudi Arabia produces just over $1.2 trillion. We have crossed into a place where a single private business, on its first trading day, will be worth more than the annual economic output of every G20 country except a handful at the top.

GDP was not built to measure this. It was designed in the 1930s by Simon Kuznets to compare national productive capacities in a world where the largest companies were a small fraction of the economies they sat inside. By 1953, at the peak of its postwar dominance, General Motors accounted for around 3% of US GDP. SpaceX, on its first trading day, will be worth roughly 6 to 7% of the entire US economy, all by itself.

When the largest single units of economic value stop being countries and start being privately controlled companies, you cannot describe what's happening with a country-level number. The unit of measurement is wrong. It would be like trying to measure ocean depth with a yardstick. You can put a number on what you can reach, but you are not measuring what you think you are measuring.

Where The Gains Actually Went

GDP doesn't tell you any of this.

US GDP has roughly tripled since 1989, in real terms. Over the same period, the share of US household wealth owned by the richest 1% rose from 22.8% to 30.8%. The bottom 50%, all 66 million households of them, ended 2024 holding 2.5% of total US wealth between them. The richest 0.1%, about 130,000 households, quintupled their wealth from $4.45 trillion to $22.48 trillion in those thirty-five years. The 905 billionaires the United States now contains hold a combined $7.8 trillion.

You can stack those numbers however you want. Top 10% owns 70% of everything. Top 1% owns 31%. Bottom 50% owns 2.5%. Pick the cut. The pattern is the same regardless. Wealth in this country has concentrated, at a pace and to a degree not seen in a century, while GDP marched cheerfully upward, telling the public a story of growth that was, in any real sense, only happening for the people at the top.

And we haven't even talked about where the money the workers didn't get went. That's the next part.

Stock Buybacks: A Trillion Dollars A Year

When a company has more profit than it knows what to do with, there are a few things it can do. Raise wages. Invest in research. Lower prices for customers. Hire more people. Build new facilities. Or, increasingly, it can use that money to buy its own shares back from the market, reducing the share count, mechanically increasing the value of every remaining share, and rewarding the people who already own them.

The fourth option, the buyback, used to carry real legal risk. The Securities and Exchange Commission treated it as potentially manipulative under its anti-fraud rules until November 1982, when SEC Rule 10b-18 was adopted and gave companies a safe harbor to repurchase their own shares without triggering manipulation liability, provided they followed certain conditions on timing, price, and volume. The timing of that change, alongside everything else we've talked about, is not a coincidence.

In 2024, S&P 500 companies spent $942.5 billion on stock buybacks. That was a record. In 2023, they spent $795.2 billion. In 2022, they hit $922.7 billion. In 2021, $881.7 billion. Three of the four most recent years, US public companies have spent close to a trillion dollars apiece buying back their own shares.

Apple, by itself, spent $104.2 billion on dividends and buybacks in 2024. Up 24% from the previous year. The combined buyback and dividend total returned to shareholders in 2024 was $1.572 trillion. Trillion. From the S&P 500 alone.

A trillion and a half dollars a year, every year, going directly to the people who own the equity, instead of to the people who do the work, instead of to research, instead of to lower prices, instead of to expanded operations. And every year, the GDP number ticks upward in a way that obscures completely where that money is actually flowing.

The buyback machine is the most efficient mechanism for upward wealth transfer ever invented. It is technically legal, technically efficient, and it has worked exactly as designed for forty-three years.

There is an argument you'll hear at this point, usually from people who have been comfortable for a long time, that all of this concentration is a feature rather than a bug. That the rich getting richer is the price of dynamism. That as long as the absolute floor for the poorest is rising, the relative gap doesn't matter. That envy is unattractive and progress is unstoppable and on it goes.

There's a kernel of truth in it. Absolute poverty in the United States is, by most measures, lower than it was sixty years ago. The poor of 2026 have access to communications technology and medical capacity that the rich of 1960 could not have purchased at any price. That is real, and it would be dishonest to pretend otherwise.

But the rest of the argument falls apart on contact with the data. The American floor has not actually risen for the median worker since 1979. The productivity-pay gap, the wealth-share data, the stock-ownership concentration, the buyback flows, all of it tells the same story. The poor are not appreciably better off in real terms than their parents were. The middle has fallen behind in any meaningful comparison. Healthcare costs, education costs, and housing costs have grown faster than wages for forty-five years, and any honest accounting of "you're better off now" has to grapple with that.

The deeper problem is this. When wealth concentrates this tightly, it stops being just wealth. It becomes political power, regulatory capture, control over the institutions that decide whose interests the economy serves next. A small number of people, owning a vastly disproportionate share of the country's productive capacity, will spend some of that capacity ensuring the rules don't change. That isn't conspiracy. It's just what people with power do, in every century, in every country. They preserve the conditions that made them powerful.

GDP keeps that conversation off the table. It says the country is doing fine. It doesn't ask who "the country" is.

This is what economists and politicians have actually been doing for forty years when they quote GDP at you. They are not lying, technically. They are pointing at a real number. But they are pointing at the one number, out of all the available numbers, that is structurally incapable of telling you the story you most need to hear. They are doing it on purpose, or out of habit, or because they were trained that way, or because the people who pay their salaries prefer the conversation that GDP enables. The motive varies. The effect is consistent.

A single number cannot describe a complex economy. The longer we pretend it can, the longer the people who benefit from the gap between the headline and the reality get to keep operating in the dark.

A few things you can actually do with all of this. Not to fix it on your own, because you can't. To see past the headline.

When someone quotes you GDP, ask the next question. What did median household income do over the same period? What did wages do compared to productivity? What did wealth concentration do? What did stock ownership do? Every single one of those follow-up numbers is publicly available. The Bureau of Labor Statistics publishes wage data. The Federal Reserve publishes the Distributional Financial Accounts, which break wealth ownership down by percentile in granular detail. The Economic Policy Institute publishes the wage-productivity gap every year. None of these require an economics degree to read. All of them tell a fuller story than the GDP headline ever will.

Learn the actual numbers. Top 1% owns 30.8% of US wealth. Bottom 50% owns 2.5%. Median full-time worker wage was about $49,500 in 2024. CEO-to-worker ratio at S&P 500 firms is around 285 to 1. Stock buybacks ran at $942 billion last year. These are not hard to memorize. They are easy to deploy in conversation. Most people you talk to about this will have never seen a single one of them.

Be skeptical of every "the economy is doing well" claim that doesn't say which economy and whose. Every claim of that shape is selecting one number and ignoring ten others. The habit to hold yourself to from here on is asking which number, and whose share, before you accept any sentence about how things are going.

And vote, and argue, and push back on this with full information. The structures that produced a $1.75 trillion private company alongside 66 million households scraping at 2.5% of national wealth are not natural laws. They were built by policy choices, mostly in the last forty-five years, mostly starting in 1979. Policy choices can be remade. They probably won't be remade quickly. They certainly won't be remade by any one person. But the conversation starts with people actually seeing what's happening, in plain numbers, without the GDP-headline blanket thrown over the top.

The era of one number describing the health of an economy is over. It probably should have ended decades ago, when Stiglitz's commission wrote its report. SpaceX is just the moment that made it impossible to keep pretending. A single company is about to be worth almost as much as the entire annual output of Spain. That isn't a statistic about a great company. It's a statistic about a system that has stopped distributing the fruits of growth in any recognizable way, and a measurement system that was built to hide exactly that.

Next, on Thursday, who actually ends up wealthy in a world built like this. And who doesn't.

More to come after that, on what to actually do with this picture. In your work. With the tools. Inside an economy shaped this way.

Best

Jono