Byrne's Blog

The Economics of Advertising: Why Advertising Agencies Used to Be the Best Business in the World (And Why They Never Will Be Again)

There’s only one notably successful business personality who made his money in the ad agency business. He’s an accountant, and that should tell you something. The ad business is simply not a great place for making money.

Certainly, money has been made from advertising—Rupert Murdoch and Mark Zuckerberg both made their billions from running ads, and well-marketed products like Nike, Windows, and Walmart have produced their share of rich people.

But it’s been a while since anyone started, joined, or bought an ad agency expecting to make a phenomenal amount of money. That wasn’t always the case, though. Albert Lasker was able to throw around many millions of dollars undoing the damage some of his ads did; his employee Claude Hopkins ended up making a salary close to $5 million (inflation-adjusted) for writing copy; David Ogilvy went into advertising because he knew it would make him rich; and the CEO of Ted Bates got a $100 million payday when the firm was bought out in 1986, at a time when that was still serious money.

There’s a good reason these people got Wall Street-sized compensation:for decades, ad agencies charged their clients in a way very similar to how hedge funds do now—but, for reasons I’ll get into, the agencies themselves were even more profitable.

“A royalty on the growth of others…”

Warren Buffett describes the ideal business as one that, among other things, earns its profits as “a royalty on the growth of others,” requiring little extra investment. He also uses the “tollbooth” metaphor: once you’re there, you keep collecting revenue without spending much. The math works amazingly well for an ad agency following the traditional model. Let’s assume that in Year One of our study, the agency handles ten million dollars in ad business. That gives it revenues of $1.5 million. Assume that, at this level, the company’s net profit margin is 5%.

What happens five years later? Their revenue probably grows as fast as that of their clients—say, 7% per year. But their expenses might only rise at the rate of inflation. Call that 3%. After five years, revenues are $2.1 million. Their expenses are $1.65 million. That leaves about $450K in profits, for an annual growth rate of 43%! What a tollbooth!

That’s good for a small business. It’s great for a business that didn’t have to make any major decisions, go after any new clients, or do anything it wasn’t doing five years earlier. No wonder Buffett liked these companies.

It gets better, though. Forget looking at a small agency: what about a big, successful one? Let’s assume they’re very good at what they do: they attract faster-growing clients (+1% growth), they do a good job growing their clients’ sales (+1% growth), and they can convince their clients to spend more money (+1% growth). Not only that, but their success attracts new clients—let’s call that 5% more growth. Now, those new clients will require someone to actually do the work, but those people will be working for a big, secure, growing company; it won’t be hard to attract that right people, at a reasonable price. So expenses can grow 4% faster than in the small company example. Since we’ve already established that past growth leads to high profits, let’s start this company with a $100 million in billings, $15 million in revenue, and a 20% profit margin.

What happens after five years? Their revenues double to $30 million. Expenses rise to $16.8 million. But that leaves $13 million for profits—growth of 32% per year. It gets better: with such high profit margins, this kind of company can better survive a recession. A 10% drop in revenue will hurt their profits, but it will wipe out smaller agencies. So the big agency can also grow by scooping up competitors. Buffett was serious about buying this kind of stock (in the 70’s, tollbooth-style ad agencies started showing up in his portfolio. Along with stock in an actual toll road company, Ogilvy & Mather was a big Buffett holding in the 70’s and 80’s).

What Happened?

Clients got sick of it and started asking for hourly rates. It’s a bit more complicated than that (the big money played a role; clients eventually realized that they were subsidizing three martini lunches; and once the cartel-lite pricing behavior of the big agencies weakened, there was no good reason to be part of the cartel).

And that completely changed the economics of the business. It’s not a royalty on the growth of others: it’s hour-arbitrage: buy forty-hour weeks from writers, designers, etc., and sell those hours to clients. If you can sell all forty (or fifty, or sixty), you will do very well, indeed. If you can only sell thirty, you’ll break even. And if you can only sell twenty-five, you’ll lose money.

But this also makes it insanely easy to scale up or scale down. Got more clients? Buy more hours. And you can subcontract, too: sell hours to a big client, buy some of your hours from a smaller, more specialized company.

Instead of a series of high-margin, high growth, low-risk streams of cash, the business has basically turned into a way to aggregate the work of free agents.

So Are Hedge Funds the New Best Business in the World?

If you want a royalty on the growth of others, running a hedge fund seems to be the way to go. Under the normal compensation structure, you can earn 2% of assets and 20% of profits. On a very mediocre 10% return, that means 4% of assets go to the managers each year. And if that 4% of assets can buy good analysts, savvy traders, and the like, one could expect even higher returns on even more capital.

But it’s not that easy, precisely because it is so easy. In finance, people with quantifiable value tend to get paid extremely well. So if your analyst has a knack for picking good energy stocks, he’ll get a great job offer from the biggest energy investor who hears about him; if your trader has an awesome talent for dealing with distressed debt, he’ll get hired by anyone with a big portfolio of the stuff. The only way to reduce turnover is to pay such vast amounts that you don’t risk losing our people.

Advertising doesn’t quite suffer from the same problem. It takes lots of creative people to put together a single campaign, and it’s very hard to say that the copywriter contributed X% of the profits, while the art director was responsible for Y. Direct response is different, of course, but it scales even worse for companies, since anyone getting paid the salary they’re worth is someone who could be making far more money as a freelancer.

While the top hedge fund guys are making a lot of money, they aren’t making money from the business advantages of the hedge fund structure; they’re making money because the measurable results allow them to take a significant fraction of the money they earn for their firms. This is why the richest hedge fund owners have privately-held companies, and the publicly traded hedge funds have been such a disaster. They’re good to work for, but only good to own in the sense that you can work for them.

What’s the Best Business In the World Now?

If advertising agencies are no longer the best business in the world, and hedge funds never will be, what is the best business in the world?

It’s Facebook. Moore’s Law means that their costs approach zero. Metcalfe’s Law means that their revenues approach infinity. Everything else is building and operating the Buffett-style “tollbooths” to collect the money.

It could have been Google, but Google has trouble earning a marginal profit on the data they collect outside of search. Google’s search/Adwords business is great, but not if its returns are being invested in the rest of Google, which is basically the world’s biggest VC fund.