Should households be granted the right to control their personal information and to refuse to give it out, as some privacy advocates insist? Or are those economists right who argue that privacy in any form is harmful since it restricts information flow and hence inhibits decision-making, increases transaction costs and encourages fraud?

Two professors at University of California, Berkeley's Haas School of Business have recently weighed in on this seemingly endless debate to argue their conclusion that neither approach is right.

In an article in the September issue of the journal Quantitative Marketing and Economics, Professors Benjamin Hermalin and Michael Katz note that privacy can be efficient in certain circumstances but that privacy property rights - personal control over one's personal information - are often worthless.

"Our analysis demonstrates that there are complicated tradeoffs missed by both sides of the debate," they write. "Certainly in the case of employment, changes in privacy policy can make some households winners and others losers."

The authors note that there has been a long history of contentious policy debates and governmental efforts to protect personal privacy, particularly the ability to maintain control over the dissemination of personally identifiable data: privacy as secrecy.

And they say recent technological developments in information collection and processing have heightened privacy concerns, with online bookstores knowing what you like to read, TiVo reporting personal viewing habits to the company's central database, and airlines keeping a record of where you travel. Meanwhile every year privacy bills are introduced in state legislatures and the US Congress in response to privacy concerns, yet there is little consensus on the appropriate approach.

"There are many calls for strong governmental intervention to restrict the use of personally identifiable data. However, there are also calls simply to establish appropriate property rights to information on the grounds that market forces will then lead to efficient privacy levels," they say.

The authors note that proponents of the Chicago School have labelled privacy harmful to efficiency because it stops information flows that would otherwise lead to improved levels of economic exchange. And they agree there are some situations in which allowing households to reveal personally identifiable information is beneficial because it allows firms to make tailored offers that facilitate transactions that otherwise might not have occurred.

Yet they insist that, contrary to the Chicago School argument, the flow of information from one trading partner to the other can reduce ex post trade efficiency when the increase in information does not lead to symmetrically or fully informed parties.

With so many people making extreme claims in discussions of privacy and related public policy, and with so little understanding of the underlying economics, it is important to identify the fundamental forces clearly, they conclude.

"Both sides of the e-commerce privacy debate have overstated their cases," they say.

While failing to come to any definitive conclusions about whether one can identify conditions under which public policy should or should not promote privacy, they authors conclude that the assignment of privacy rights to personally identifiable information may have no effect on agents' equilibrium welfare levels and need not lead to an efficient equilibrium privacy level.

"In some situations, the only effective policy would be explicitly to block the dissemination or use of such information. Public policy could block dissemination in several ways. One is to make it illegal to reveal personally identifiable data. Another is to destroy employment or prison records or other forms of tangible evidence, which would prevent households from credibly revealing the information even if they chose to do so. A related policy would be to refuse to enforce sanctions against people who lie about their protected characteristics," they conclude.