A Cash-Free Society

This chapter follows three historical and parallel developments. The first focuses on the development of alternate means of payments other than cash, such as checks and credit cards, as a run-up to the digital payment systems. In the second we shall look at standards, such as product codes and bar codes. Interestingly, these standards, which are so important for a digital society, came many years before the Internet and the www technologies that really made use of them. The third trend is the development of computer technology. When these three trends—credit cards, standardization, and technology—merge we have the complete foundation for a cash-free society.

In the analog world—that is, before the advent of digital computers—paper was the most important medium. Cash was on paper and the early replacements of cash were also paper based. Even the first credit cards relied on paper. While the card itself could be plastic, the imprint of the transaction was made on a paper slip. While paper has many advantages (for example, anything can be printed on a piece of paper), the disadvantage is that all further handling must be physical. Checks and credit card slips have to be transferred physically to the banks, where they must be counted physically. There were some possibilities of using machines in the processing, but these were limited by the fact that much of the data was in handwriting. Later on, as computer scanners and optical character recognition became more advanced, some of these processes could be automated.

In many countries, paper is now out of the payment loop. Checks are only used rarely and all credit card transactions are digital. However, it is still interesting to take a look at these “early technologies” as a means of explaining the development toward a digital society. We will also see that a set of standards are necessary when formalizing the payment systems. Then we have the background for using the development of computer technology to automate the complete payment system.

A check is a letter asking a bank to pay a specific amount to the bearer of the check, deducting the amount from the account owned by the person who wrote the check. Checks have been used for more than a thousand years. In fact, early versions of checks were in use by the Romans. The first preprinted check forms were issued by the Bank of England in 1717. The Bank of New York issued the first checks in America in 1784.

The use of checks peaked in the second half of the twentieth century. Instead of paying employees in cash, which had been the norm, banks persuaded companies that it was simpler to pay using checks. A clear advantage was that one no longer had to operate with large amounts in cash, transporting cash from the bank to the company, counting the money for each employee, and requiring the employee to come to the office to collect the salary. On their side, employees could now take their check to the bank and withdraw the whole sum, but banks persuaded customers to let the money stay in an account and to use their own personal checks for large expenses.

The new system of using checks was far more efficient and more secure than the old system based on cash. Employees also had the advantage of not having to handle large cash amounts. Psychologically it became easier to save a part of each month’s salary, just by letting it stay in the bank. Checks were more secure than cash, but part of the expenses of such a system had to be paid in transaction fees.

The banks were the winners. Checks enabled them to be a part of the weekly or monthly salary transactions. Most employees of firms that paid with checks became their customers. After some years it became the norm to pay salaries using checks, even for many smaller firms. Then all participants in the workplace became bank customers. Similarly, since checks were used to pay in shops, restaurants, hotels, and elsewhere, most merchants also became bank customers. For them it was also convenient to endorse the checks and insert the money into their own account, rather than withdraw it as cash.

The problem with checks is that they can be misused. The techniques used to avoid falsified checks were the same as those used to protect paper money, such as special paper, making it difficult to alter the data after the check has been written, and watermarks. A clear weakness with checks is that they are authorized with only a signature, and that the receiver will not know if funds are available before the check is cleared.

Because checks are in paper form, they must be transported physically to the bank. Handling checks became easier when the bottom part of the check was imprinted with machine-readable characters. This opened the way for using machines to sort checks. Data other than the preprinted information, such as date and amount, had to be typed in. Later, programs were used that could recognize the hand-written numbers. Still, the physical and manual parts of check handling make it a costly form of payment. Bank of America estimates that handling a business check can cost between $4 and $20 if one includes all costs, such as the price of the check, the time used to write the check, mailing, and so on.¹ It is estimated that checks cost 10 times more than electronic payments.²

In clearing checks, banks are now moving images of checks electronically rather than transporting the actual paper. It has been estimated that this method saves banks more than a billion dollars every year.³ In addition, consumers and businesses benefit from faster processing.

Credit cards, in their different forms, have a long history. They were originally issued by banks and other companies to frequent customers and often used to pay at gas stations. The idea was that static information, such as the name of the customer and account data, was embossed on the card in such a way that the card could be used to make an imprint. As the name implies, the basic idea is shop now, pay later.

American Airlines introduced an Air Travel Card as early as 1934. These cards had numbers that identified the customer and the account. It took just a few years before most major airlines accepted the card. Soon, half of the ticket revenues came through these cards. A breakthrough came when the same card could be used to pay many different merchants. The first general purpose cards were Diners, followed soon by the BankAmerica card.

Bank of America faced the same problem that we have today when introducing a new technology—that is, the advantage of most so-called network systems increase as more users are on. We see this today with social systems—Twitter, Facebook, LinkedIn, WhatsApp are good examples.⁴ The challenge is then to get users, initially forgetting about revenue. Bank of America solved this by issuing credit cards to all its customers in Fresno, California—60,000 in all. Having done this, the bank could persuade merchants to accept the card.

While the impression from the card took care of the static information, the amount was entered manually on the slip and was then signed by the customer. In order to verify the card and the transaction, the merchant would have to call the bank or the credit card company. Here one had to perform a manual control, sometimes certifying the credit balance of the customer. Once the transaction had been found to be in order, the merchant would in some cases write a verification code on the slip. Then the slip would have to be sent to the credit card company or, in many cases, to a company that specialized in handling credit card transactions.

The idea behind the general purpose credit cards, as well as with the checks, was to get the banks and the credit card companies into the payment loop. While a cash transaction could go directly between the buyer and seller, without involving any bank, a credit card or a check placed the bank in a central position. A clear incentive for the banks was that this offered an opportunity to offer credit with high interest rates. While some customers managed to pay the outstanding bill every month, there were many who could not withstand the idea of shop now pay later.

Advances in computer technology promised a simpler future. Starting in 1973, Visa developed a computerized verification system that significantly reduced the time it took to verify a transaction. The merchant still had to call Visa, but the answer would be given in less than a minute. The paper slip was still in use, but as we shall see in Chapter 8, credit cards could easily be transformed to a digital economy.

Today, many credit cards offer a 30 to 45-day interest-free period, after which interest has to be paid on the total debt. Most cards require that part of the outstanding debt is paid every month. Credit card interest rates are often much higher than the interest most customers have to pay on regular bank loans. These expensive loans offer a good profit opportunity for credit card companies.

Credit card issuers may offer additional benefits, such as discounts or including travel insurance. Some cards offer a cash-back system, where customers receive a small percentage of their purchases back, either as cash or as bonus points. The retailer accepting the card payment will have to pay a fee, often between 1.5 and 3 percent of the purchase amount, or even higher for some cards. While credit cards have many advantages for customers, at least for customers that do not have a large outstanding debt on the card, the retailer has been burdened with the costs. As credit cards become the norm, it is very difficult for a retailer not to accept a card. This is especially the case for the hospitality business, such as restaurants and hotels.

Technology has made it easier for companies to offer credit cards. The universal cards are still dominated by the incumbents, such as Visa, MasterCard, Diners, and American Express. However, large stores offer their own store cards that can only be used within that particular chain or store. The advantage for the store is that they get the full amount of what the customer pays. A store card is also a convenient infrastructure for offering discounts and building customer loyalty.

Before the advent of self-service groceries, milk was sold from churns; grain, sugar, and potato from sacks; and tea from boxes. Just a few items were pre-packed. Home deliveries, which are today presented as a new and disruptive service, were also available many years ago. In several countries one could have milk delivered to one’s door, as well as bread and some other products. There were trucks driving around the neighborhood offering fruit, vegetables, and even fish and meat.

With the advent of the self-service store, the need for packaged products became greater, as this allowed for greater efficiency. Pre-packed products also had the advantage that one could use price tags, which allowed for greater efficiency at the cash register. However, the drawback of just typing in the price here was that one had no information about what was being sold; there was no data for providing a receipt to the customer (except for the amounts), no data for maintaining inventory, and no data for statistics.

The first means of identifying a product came as early as 1948 with a system that could automatically read product information during checkout. The idea of stretching the dots and dashes of Morse code into bars with different thickness came immediately, but it took many years before equipment could be developed that could read these codes in an efficient way.

Parallel with the computer readable code was the development of a universal product code (UPC). It was clear that the success of these two developments would come when they could merge, when producers all over the world could put a unique UPC on their products as a computer-readable bar code that could be scanned automatically. The next step came when the data from the bar code could be used to do online lookups in computer systems, to retrieve the price of the product, to register the sale for statistics, and to do a count-down on stocks.

This is a very good example of the symbiosis between technology—represented here as bar code readers and databases with prices and product information—and standards, here the UPC system. Later on we shall see that additional technology, such as data networks, are needed to realize efficient digital payment systems.

Today, most products have an International Article Number, otherwise known as a European Article Number (EAN). Other national codes, such as the American UPC, have been embedded into the EAN. The main part of an EAN offers information on the manufacturer and product. The ISBN, the International Standard Book Number, is now compatible with the EAN. An ISBN identifies the country, publisher, and a unique number that the publisher or some other agency assigns to the book. This can be, for example, a consecutive numbering of books from 1 upward for this publisher.

Lasers are now used to scan bar codes. The devices are inexpensive and robust and the process of scanning is so simple that it is often left to the customer—for example, in self-checkout machines, a growing trend in department stores. These can scan bar codes from all directions, which leaves the customer only having to put items on a conveyor belt.

Printing bar codes is a simple task. The EAN-13 code can be found as a font that can be installed in a standard word processor, such as Microsoft Word. Numbers can then be printed in the bar code font, as seen in Figure 6.1.

In some settings, two-dimensional QR codes (as shown in Figure 6.2) are used. These can code more information than a bar code and need somewhat more sophisticated scanners. Some smartphone apps can read QR codes by taking a picture of the code. Most often the QR code offers a link to a website, for example to present information or to run a process.

Radio-frequency identification (RFID) is another possibility. RFID tags may be passive or active. Passive tags are inexpensive but need to be close to the reader. Active tags have a battery in the circuit and still cost just a few cents. The advantage is that the reader no longer has to be in direct or close contact, with the tag. With RFID, one could theoretically carry a bag of groceries out the door while a scanner reads your credit card and the price of all items. In practice, it is not as simple. Even with battery-powered tags it is complicated to read items stored in a bag or in a shopping cart. However, RFID is used in cards for contactless payments (along with other technologies), in ID cards, and many other applications.

The previous section on product codes and bar codes provides a good example of automation. In order to have an efficient check-out function, every product needs a unique code. Manually identifying products by type or name is time-consuming, spelling may pose a problem and the same product may exist under several different types and names. All of these problems are avoided by having a unique product number, leaving it to the producer to decide if a new version of a product should get a new number or not.

In order to get all producers all over the world to adhere to the code scheme, it must have some clear advantages. These come when the code can be presented in computer-readable form, in our example as a bar code, and where good scanners are available. Then the stores install scanners, and then most producers put bar codes on their products and there will soon be a situation where stores will only sell products that have a code. A more formalized world is created.

As we have seen, technology is an important factor. Bar code scanners need to be inexpensive and robust. To achieve efficiency, the code must be scanned in a split second, keeping errors to a minimum. Many of the early bar code systems were abandoned because the readers, which often used incandescent light bulbs, were unreliable. Reliable scanning systems only existed when these were replaced with lasers.

All of this would have little effect on its own if the scanners could not be connected to other systems. A manual cash register would have little use for a scanner. In the beginning, computers were too expensive and too large to incorporate in a point-of-sale terminal.

In the early 1970s, Intel and other manufacturers developed a new way of making computer components. Instead of using separate electrical components, such as transistors and capacitors, and connecting these with wires, one could print the components on silicon chips. Large-scale integration (LSI), and later very-large-scale integration (VLSI), became the norm. In practice this allowed for the production of inexpensive, small, robust, and energy-efficient computer processors, the backbone of PCs, laptops, smartphones, and any computer device. Disk technology had a similar evolution. In a magnetic disk, information—in practice, binary numbers consisting of zeros and ones—can be stored on the disk by magnetizing the surface of the disk. From large units with minimal capacity in the 1960s, disk development mirrored the development of the computer; each new version had more capacity and was smaller than the last.

Now the cash register could be a computer. The scanned product code could then be transferred to the computer and used for a lookup in a database, where the name of the product and the price can be found. This could then be printed on a receipt given to the customer. The cash register displays the total. If the customer pays digitally, for example by offering a credit card to the point-of-sale terminal, all of the information needed to perform the payment is then available digitally: the credit card number, other data from the card, then the amount and date from the cash register. The cash register would then transfer all the data in a standardized format to a central clearing system.

In addition, the store will now have valuable data about what the customer bought, which would be even more valuable if the customer was identified to the store; for example, by using the store’s loyalty or credit card (see Chapter 8).

All point-of-sale terminals are now online and the payment transaction, including verification of the payment card, can also be done online. This is not a problem in a society that has a good digital infrastructure. Without an online solution, however, as was the case in the early days of networking, the transaction data could be saved in the terminal during the day and then sent to the central clearing system after shopping hours, most often over a dial-up network. This “batch” system did not allow for online verification of credit cards.

The online variant offers other advantages. For example, it may be used to offer loyalty discounts directly to customers. The system will do a lookup based on the customer ID and the items that are bought and automatically compute a discount that is withdrawn from the total amount before paying. With a good broadband connection, this may be performed immediately.

The transition from cash to representing money in a computer is clearly not much more radical than moving from using precious metals and coins to banknotes. What we see is that the concept of money can have many representations, similar to books, music, maps, and most other symbolic systems. Digital representations today have many advantages for storing, analyzing, and transporting data.

For money, there are other advantages as well. The data can be collected in electronic form where they originate, such as in a point-of-sale terminal. These data can then be transmitted immediately to credit card companies and banks in their original form without any need for human intervention. This makes it possible to have updated balances on all accounts. It is also possible to retain all data from the transaction, not just the sum but also a list of items that were bought. In the future (it is already here in some versions) there will be no need to retain the receipt from the shop as the same data will be available online.

¹ https://www.wsj.com/articles/SB10001424052702304732804579425233344430424

² http://ww2.cfo.com/applications/2015/10/electronic-payments-10-times-cheaper-checks/

³ http://www.businessinsider.com/the-death-of-the-paper-check-2013?r=US&IR=T&IR=T

⁴ Interestingly, we may also have an opposite network effect—that is, when a person has so many friends on Facebook (parents, old boyfriends, teachers, and more) and get so many posts that the only solution seems to close the account.

Chapter 6 From Analog to Digital