NBC, CBS, and the Invention of the National Advertising Market

A technology with no way to pay for itself

American radio broadcasting arrived without a business behind it. Stations went on the air through the early 1920s selling nothing but the idea of the broadcast itself: manufacturers hoped listeners would buy receivers, and beyond that no one had settled how a station was supposed to cover its own studio, staff, and power bill. There was no subscription fee paid by the listener, no license fee of the kind that funded broadcasting in some other countries, and for the first two years of American broadcasting, no advertising of any kind on the air. A station was, in effect, a promotional expense for whoever built the transmitter, not a business in its own right.

AT&T supplied the fix, through a station it operated in New York under a plan its own engineers named toll broadcasting: sell time on the air the way the company already sold long-distance telephone lines, by the minute, to whoever wanted to rent it. The logic was borrowed directly from the phone business AT&T already understood. A subscriber paid for the use of a wire for a stretch of time; a sponsor, under the new plan, would pay for the use of a transmitter for a stretch of time. On August 28, 1922, that station, WEAF, aired what is widely regarded as the first paid radio commercial, a ten-minute spot for the Queensboro Corporation's Hawthorne Court Apartments, sold for $50.

The scale in 1922 was small enough to read as a curiosity rather than the birth of an industry. By October, WEAF had logged total time sales of just $550 under the toll system, a handful of sponsors testing a format nobody yet trusted with real money. What the toll system did establish was a working mechanism: a station could sell a slot, a sponsor could buy one, and a spoken message could reach whoever owned a receiver within range of the transmitter. What it did not yet have was reach beyond a single city, and without reach beyond a single city there was no reason for a national advertiser, the kind spending real money on magazines and newspapers already, to pay a national price for radio time.

Toll broadcasting solved one problem and immediately exposed the next. AT&T's plan proved a sponsor would pay real money for a slice of a broadcast day, and that a station could sell that slice without pretending it was anything other than paid time. What the plan could not do on its own was connect the New York studio to any other city, so a listener in Boston or Philadelphia sat entirely outside the transaction, invisible to WEAF's sponsors no matter how badly an advertiser might have wanted to reach that market too. The gap between a station that could sell a slot and a system that could sell a country was the gap RCA moved to close within four years.

RCA builds the network AT&T would not

The network able to deliver a genuinely national audience came from a different company than the one that had proven advertising could work on the air at all. In 1926, the Radio Corporation of America founded the National Broadcasting Company as the first American network built to link stations across the country and feed them a shared schedule of programming, and of advertising, from a single New York source. General Electric, Westinghouse, and AT&T held ownership stakes alongside RCA, folding the era's dominant electronics manufacturers and the dominant telephone monopoly into a single broadcasting company with a national reach none of them could have built alone.

A year later, in 1927, a rival network organized in New York under the name Columbia Broadcasting System, built specifically to compete with NBC for affiliated stations, for sponsors, and for listeners. Within five years of AT&T's small trial on a single New York station, the country had two organizations with the technical and financial means to feed identical programming, and identical advertising, to dozens of affiliates at once, several time zones apart.

Both networks ran on the same underlying economics. A local station signed on as an affiliate, agreed to carry network programming at set hours, and split the advertising revenue that came bundled with it. A national sponsor, in turn, no longer had to negotiate separately with stations in Chicago, Dallas, and Seattle to reach a national audience. It could buy the NBC or CBS network in a single transaction, at a single national rate, and reach the full affiliate list the following Tuesday evening, an efficiency no newspaper chain of the period could match at any price.

The money moved: radio against the newspaper

The economic case for having built the network showed up first in the industry's share of total advertising spending. Radio carried roughly 2 percent of all US advertising expenditure in 1928. By 1932, in the middle of the Depression, that share had grown to nearly 11 percent, according to a Library of Congress research guide on consumer advertising during the Great Depression. The gain came at the direct expense of the medium radio was displacing: newspapers, which had held the advertiser's dollar by default since before radio existed, watched the same agencies that had bought column inches for a generation begin buying network minutes instead, and did so while total advertising budgets across the economy were shrinking, not growing.

The gain concentrated further once it reached radio itself. Of every dollar spent on radio advertising in this period, about 60 percent went to the national networks, chiefly NBC's Red service and CBS, together with the stations the networks owned outright. The other 40 percent was split across the country's independent regional and local stations, hundreds of them, competing for what remained. Two corporations headquartered a few blocks apart in Manhattan had positioned themselves to take the majority of a fast-growing pool of money, while thousands of independent broadcasters divided a shrinking remainder among themselves, a pattern that looked less like healthy competition inside one industry and more like two firms absorbing an entire new revenue stream as it formed.

The profit figures make the transfer concrete, though they carry more uncertainty than the spending-share numbers, since they survive mainly in period industry accounts rather than audited public filings, and should be read as estimates rather than a settled census. By the same Library of Congress research guide, NBC posted a profit of about $1 million in 1932 and CBS about $1.6 million, in the depth of a Depression that was closing newspapers and bankrupting local broadcasters across the country. Two years later, the combined profits of the two networks are put at roughly $72 million. Taken cautiously, those figures still describe an industry moving real money toward two addresses in New York while most of the American economy moved money nowhere at all.

None of this happened because advertisers had decided, on some abstract measure of persuasion, that a spoken commercial worked better than a printed one. It happened because two companies had built a single piece of infrastructure, the affiliate network, that let an agency buy simultaneous national reach in one transaction instead of assembling it market by market through a hundred separate newspaper insertion orders. Buying convenience, not audience quality, is what moved the money, and it moved fastest during the very years, 1929 through 1933, when agencies had the least room to experiment and the strongest incentive to buy the format that worked with the fewest separate purchases.

One signal, the whole country, at once

What made the concentration of advertising dollars matter beyond the balance sheets of two companies was the size of the audience sitting behind it. Radio's household penetration in the United States rose from roughly 1 percent in 1923 to a majority of households by 1931, and to about 75 percent by 1937, the fastest a communications technology had reached that share of American homes to that point. NBC and CBS were not simply the two largest sellers of advertising time in the country. Within a decade of their founding, they were the only two organizations capable of putting an identical sentence into most of the country's living rooms in the same minute.

No earlier American medium could do that. A newspaper reached a city, at best a region, and reached it hours or days after an event, in whichever version a local editor chose to run. A national network could originate a program in a New York studio and deliver it, unedited and simultaneous, to an affiliate list running from Maine to California. The federal government had licensed the individual stations that made this possible, station by station, largely to sort out interference between transmitters crowding the same frequencies. It had not, in that licensing, planned for two private New York corporations to end up deciding what most of the country heard and when, years before Washington built any real apparatus for thinking about that kind of concentration as more than a technical problem of spectrum interference.

The reach kept compounding as the network system matured through the 1930s and into the 1940s. A 1947 survey by the ratings firm C. E. Hooper found that 82 of every 100 Americans were regular radio listeners, evidence of how close to total the household reach of the network model had become by the end of what broadcasters themselves called the medium's golden age. By then the pattern the two networks had set in the 1920s, national origination, simultaneous delivery, paid for entirely by advertising, had defined American mass communication for a generation, and would carry over with little change into the television networks NBC and CBS went on to build from the same New York offices.

Congress had already tried once to catch up. The Radio Act of 1927 created the Federal Radio Commission specifically to referee interference between stations crowding the same frequencies, a genuinely urgent technical problem in the chaotic early years of broadcasting. The Act said nothing about network ownership, national advertising rates, or what it might mean for two companies to control the schedule reaching most of the country's living rooms, because in 1927 that particular concentration had not yet fully happened. By the time it had, in the early 1930s, the two networks already had the affiliates, the advertisers, and the audience locked in, and a rewrite of federal communications law was still years away.

What the duopoly built, and what it cost

The case for the network system is straightforward and largely uncontested. It gave a country that was still mostly rural in the 1920s a shared national culture for the first time: the same comedy program, the same dance orchestra, the same evening news bulletin, heard at the same hour whether the household sat in Ohio or Oregon. Within a decade of CBS's founding, the medium carried a sitting president's voice directly into tens of millions of homes with no newspaper editor standing between the speaker and the listener, a use of the network's reach that a president would put to deliberate political use before the decade was out.

The case against the system did not require hindsight; it was visible from inside the decade to anyone paying attention. The same infrastructure that let radio unify the country's evenings gave two corporations, both headquartered in New York and each under the personal direction of one executive, working control over what most of the country heard and who could afford to reach it as an advertiser. There was no competing scale to check either network. A business, a candidate, or an idea that neither network chose to carry had, in practical terms, no national voice at all, and the choice of what to carry sat with Sarnoff, Paley, and the advertising agencies able to pay network rates, not with any public process or elected body.

Put side by side, the two effects are not contradictions waiting to be resolved so much as two faces of the same infrastructure. The system that let a farmer in Nebraska hear the same orchestra as a stockbroker in Manhattan is the identical system that let two New York executives decide which orchestra, which comedian, and which advertiser earned that reach. Liberation and concentration were not sequential phases of the radio era, one arriving after the other. They were the same wire, running both ways at once, from the moment the first network signed its first affiliate.

The specific technology is a century removed from anything running today, but the shape of the problem has not changed as much as the hardware has. A market with one or two gateways to a whole population's attention will send its money and its influence through those gateways, whatever machinery happens to sit behind them. The current version of that gateway is not a network schedule; it is an answer engine deciding, in the moment before it responds, which business gets named and which does not. NBC and CBS earned their concentration by building the pipes first and letting regulators catch up years later. Whether today's businesses get the same years of head start to notice who controls the answer is the open question this earlier chapter does not settle, only rhyme with.