My father thought the stock market was a scam. He was almost right.

Opinion. This is an essay by the publisher, not an analysis and not a recommendation. Where I cite numbers, the source is attached. The rest is my view, and you are free to think it wrong.

The verdict was passed in the nineties

My father held commercial power of attorney at one of the country’s largest daily newspapers. Through him, as a child, I met the first people who promised quick money. They were the same types who promise it today. Different suits, better graphics, identical sentences.

He rejected that way of earning money. His objection was not moral. It was commercial, and it fit into one sentence: That is not a real economy. That is a parallel one.

It took me thirty years to understand how precise that sentence is — and the one place where it fails.

What he actually meant

Commercial power of attorney means commercial responsibility. You sign things that hurt when they are wrong. My father thought in a chain you can walk along: there is a product. Someone makes it. There is a spread between purchase and sale, and that spread has to pay for the work in between. At the end stands a customer who returns, or does not.

What he saw in the quick-money people had no such chain. There was a promise, an amount, and a story about why the amount would grow. The part in between — the work — was missing. To a merchant that is not a business model. It is a warning light.

He had not meant the stock exchange. He had meant the people trying to sell him the stock exchange. I did not hear that distinction as a child, and it is the entire point of this text.

Where he was wrong

Financial markets are not only a parallel world. There is a part of them as real as a factory floor.

When a company issues shares or bonds, it receives money it did not have before, and builds things with it. When millions buy and sell, something hard to replace emerges as a by-product: a price. A price tells an economy where capital is scarce and where it is lying around. And the pension my father lives on today is not stuffed under a mattress either.

Denying that part repeats the quick-money people’s mistake with the sign reversed: taking one slice of reality and declaring it the whole. My father did exactly that, and it cost him something — more on that below.

Where he was right

Now the uncomfortable part. It cannot be settled with a feeling, so I will use the numbers that exist.

US financial services grew from 4.9 to 7.9 percent of economic output between 1980 and 2007. The interesting question is not that it grew, but on what. Greenwood and Scharfstein broke it down: a sizeable portion of the growth is explained by rising asset management fees — driven by increases in the valuation of the tradable assets those fees are charged on. A second important factor came from fees associated with an expansion in household credit, particularly residential mortgages, fuelled by the development of shadow banking. The authors themselves explicitly ask whether this growth was socially beneficial — and call their own answer preliminary.

Here I have to stay clean, or I overstretch the source: the study does not say that capital formation for companies has become irrelevant. It breaks down where the increase came from — and in that breakdown, fee items carry it. That is an observation about the composition of the growth, not a causal claim by the authors, and I am not using it as one.

With a merchant’s eye, something still sticks: if your revenue grows because your customers’ assets rise in price, you have not built a better product. You hold a stake in someone else’s upswing. That is a legitimate business. It is simply a different one from the story being told.

Then the second calculation. Philippon examined 130 years of financial intermediation in the United States and arrives at an annual cost of 1.5 to 2 percent of intermediated assets, with constant returns to scale. Over that period computers, networks and high-frequency trading arrived.

And here I have to be honest, or I do the very thing I accuse others of: this measure is contested. The serious objection runs: the industry delivers more today than in 1900 — more access, more hedging, different products. Then the reference quantity is not the same, and the ratio not directly comparable.

What argues for the calculation: it is not an American peculiarity. Bazot asked the same question for Europe between 1950 and 2007 — Germany, France, the United Kingdom. There too, the unit cost of financial intermediation does not fall across the period, France excepted. After the nineties it declined in Europe while remaining stable in the United States; across the whole period it rose more in the US than in Europe. And the study finds something else that squares with Greenwood and Scharfstein: the joint rise of wealth management, credit intermediation and the securities industry through the nineties and two-thousands coincides with higher unit costs.

Two independent measurements, two continents, one finding: the thing does not get cheaper. I still do not present Philippon’s figure as proof, but as the best available estimate with known weaknesses — and as the thing nobody ever told my father, which he sensed anyway.

The cheap trick I am not using

At this point a text like this would normally roll out the really big number: the notional amounts of outstanding derivatives exceed annual world economic output many times over. You can look it up at the BIS, and it sounds like proof.

It is not. The notional amount is merely the reference figure a contract is calculated on — not the money at stake. The gross market value, what the contracts are actually worth, is a fraction of it. Confusing the two manufactures an outrage the statistics do not support.

I leave the trick alone for one reason: it is the same move used by the people my father warned me about. An enormous number, a feeling, no evidence. You cannot criticise that method and use it at the same time simply because it happens to serve your own thesis today.

What I can contribute from my own books

Economic studies are one thing. Your own kitchen is another, and here I can deliver instead of quote.

We put our own trading paths through a cost calculation and published the result: two of three paths were net negative after fees, although some were in the black gross. The fee share of gross profit came to roughly 225 and 118 percent — the friction was larger than the return. The full calculation, including its weaknesses, is in the fee truth.

Now the limits of that number, so I do not commit the sin I flagged two sections above: that is three paths, not three hundred. At n = 3 there is no statistics, no representativeness and no statement about the industry — not even a robust one about us. They are also our own books, independently unaudited; you have to take my word that the figures are right, and that is a weak argument. I cite them anyway, because they are the only first-hand contribution I can make to this question, and because they point in the uncomfortable direction: against us.

What they establish is modest and still the core: between “I was right” and “I earned” sits someone who earns on every round — regardless of whether I was right. If you do not know your friction, you do not know your result.

Why “almost”

My father was right about the people he met. They were selling a story, and he sensed it without knowing a single metric.

He was wrong in the conclusion. “Many people here lie” does not imply “everything here is a lie”. It implies only: look very closely. He turned a correct observation into an oversized rule — and the rule spared him losses while keeping him out of something he could have learned. Both belong on the same balance sheet.

I inherited both. The fascination that would not let go, although nobody around me was anywhere near that world. And his mistrust, which to this day takes apart every screenshot of a gain before I have finished looking at it. That this house dismantles fraud patterns is not a market gap. It is inherited scepticism.

The third way

I never had to choose between his two worlds. I treat financial markets like honest merchandise trade.

And because I criticised analogies above, let me be clear about what this is: a working principle, not a proof. Financial intermediation and textile production are not economically the same thing; you cannot prove anything about one from the other. What transfers are the questions a merchant asks before he signs.

A product you can explain. A margin you can recalculate. Costs you know before you pay them. And customer trust you lose if you lie.

In merchandise trade these are unremarkable. Nobody celebrates a trader for knowing his purchase prices. In online finance it is a position — and that says more about this industry than any statistic I could quote here.

My father has never seen the temple. But if he looks at it, he will recognise the handwriting: product first, story second. And if there is no story, then there is none.

How this connects to the rest of my life — the mechanic’s apprenticeship, the clothing company, and the ten years in which I understood everything and lost anyway — is in my story. More from this desk in Wealth & Mindset.