Commentary: Financial Services: The Silent Engine

The boom in the finance sector shows up everywhere but in economic statistics. And that means growth has been understated

From the stock market to the mortgage market, financial-services industries rivet the attention of Americans. Yet the financial-services boom shows up everywhere but in the economic statistics. According to the official numbers from the Bureau of Economic Analysis (BEA), the output of the broad banking sector--banks plus related industries, such as nonbank credit-card issuers and mortgage bankers--has been flat since 1990 (chart). Measured output of the mutual-funds and investment industry is up, but still only accounts for about 1% of gross domestic product.

But the stagnation shown by the official data is very misleading: Virtually every real-world measure indicates that the output of financial services has actually exploded (charts). For example, consumer credit is up by some 60% over the past eight years, as credit cards become widely accepted in such places as grocery stores. Mortgage-refinancing applications are soaring, as the introduction of automated mortgage underwriting makes refinancing far cheaper and faster. The number of transactions at automated teller machines has almost doubled since 1990, even while the number of commercial bank offices and branches has actually increased by 11%. And perhaps most telling, the financial sector, which also includes insurance and real estate, has generated almost 18% of total corporate profits in the 1990s, up from only 11% in the 1980s--hardly the mark of a stagnant industry.

PRODUCTIVITY PARADOX. The undercounting of financial-services growth is critical for understanding what is really going on in the economy. For one, it means that economic growth in recent years has been underestimated. Based on academic and government research, the true growth rate in banking could be anywhere from 5 to 10 percentage points higher than the official numbers. Assuming this applies to related industries as well, it could add as much as 0.3 percentage points, or $25 billion, to GDP each year. It would also push productivity growth higher by about the same amount.

Moreover, the flawed data on financial-services output may help explain the so-called productivity paradox--the apparent inability of computers to boost productivity. In real life, technology clearly makes a big difference. For example, Mastercard International Inc. now can handle 8.3 billion transactions annually, double the 1994 volume, using only about 20% more workers, says William I. Jacobs, executive vice-president of global resources. Financial-services firms are among the biggest high-tech users, spending more than $60 billion annually on information technology, notes Jim Moore, financial-services-industry vice-president for the Gartner Group. Yet any productivity gains from such huge investments are not picked up by official numbers.

To be sure, the government's numbers are slowly improving. The Bureau of Labor Statistics just introduced a new price index for property-and-casualty insurance premiums, which will lead to better measures of real growth in the insurance industry. Over the next couple of years, better price and real growth statistics will come out for life insurance, investment banking, and mutual funds.

The data for the banking sector, however, are still a disaster. Right now, the BEA measures output growth of the broad banking sector by the number of employee hours worked in these industries. As a result, productivity--which is output divided by hours--is effectively defined to be flat.

"INCREDIBLE DISPUTE." Unfortunately, while almost everyone agrees that productivity is up sharply, there is no agreement on how to fix the numbers. "There's still an incredible dispute about how to measure the output of a simple bank that takes deposits and makes loans," says Jack E. Triplett, a former BEA chief economist now at The Brookings Institution. Some statisticians argue that loans provide the only output from banks, since deposits by themselves don't make money. Other economists find the idea that depositors don't receive services from holding their money in a bank to be ridiculous.

Moreover, much of the technology investment of recent years has gone into providing new services and products rather than cutting prices of existing ones. For example, rather than lowering costs by replacing branches with ATMs, as many analysts predicted in the early 1990s, banks have increased the number of both branches and ATMs. The result: increased customer convenience, which has a real and substantial value--but is very difficult for statisticians to measure.


There seems to be a fundamental ambivalence about the role of financial services in the economy. If Dell Computer Corp. sells more computers, that is taken as an unambiguous positive for the economy. But if banks issue more credit cards or make more loans, there is a fear that they are taking on too much risk that could threaten the health of the economy.

While it may be prudent to worry about banks assuming too much risk, that's not a good reason to omit their enormous growth from the numbers. Financial services are vital to growth--and they should be counted that way.

Before it's here, it's on the Bloomberg Terminal.