Paul Armer, "Privacy Aspects of the Cashless and Checkless Society", April 1968

Paul Armer of RAND testified before the Senate Subcommittee on Administrative Practice and Procedure, on the topic of privacy in financial systems, in February 1968. He notes that while the controversial Federal Data Center had become politically nonviable by this time, the automation of the financial industry was proceeding rapidly, bringing its own set of privacy issues. He identifies four main factors.

Since I am not concerned here with the implications for the financial world but rather for privacy, let me focus on four attributes of the various possible future systems-- attributes important to privacy. The first is the percentage of financial transactions actually recorded; the second, the amount of detail about a transaction that is recorded and subsequently computerized; the third, the degree to which this information is centralized; the fourth, how rapidly the information is transmitted to the central computer.

Why are these particular parameters important to the privacy issue? The percentage of transactions recorded is obviously important because it determines how complete the picture is. The same is true for the amount of detail recorded about a transaction. The amount of detail actually computerized is important because if we visualize future computer systems with remote terminals connected to the computer, only the information actually in the central computer will be available electronically at the remote terminals. Information written on a piece of paper such as a credit card sales slip or check and not in the computer could only be obtained by someone eyeballing the piece of paper.

Lastly, the rapidity of transmission is important chiefly when we think of such a system being used for personal surveillance. The value of information about an individual's whereabouts declines rapidly with time.

The extreme case, in which all transactions go through the system and all the details are recorded (who, what, where, when, and how) and then sent immediately to a single center, obviously represents the greatest threat to privacy. Such a system would know where we are and what financial activities we are involved in everytime we so much as buy a candy bar or pass through a toll station. Now it is unlikely that we will get to this extreme situation in the near future, if ever. But how fast are we moving towards it, even if we may never reach that limit?

The prediction that electronic transactions could leak physical location data seems right on.

Still present is that contemporary concern with data centralization as a threat to privacy:

[C]entralization of data is usually a concomitant of computer use. The payoff to successful snooping is much greater when all the facts are stored in one place. Though most of the data to complete a dossier on every citizen already exists in the hands of the government today, it is normally so dispersed that the cost of collecting it and assembling it would be very high.

Today we have the opposite problem: we have no idea where all our data is.