From time to time I encounter an article that provides valuable insight into how computer-based automation is affecting employment throughout the world. Here is a recent example of an article about the Swedish banking industry. Quoting from (The Download, 7/30/2018):
Casper von Koskull, the CEO of Sweden based Nordea Bank, predicts that the banking industry will slice its workforce in half over the next 10 years. Last year, Koskull announced his company would cut 6,000 jobs in favor of automation.
The cuts: So far, it has cut employee numbers by 2,500. Jobs cut include those involving asset management and customer service.
The results: That’s been enough for Nordea to see huge monetary benefits so far. According to second quarter results, the bank reduced costs 11 percent year over year, and increased profits by 31 percent, which made Nordea the top performer among Sweden’s major banks.
Why it matters: Nordea’s experiment is motivating other regional banks to automate more quickly. The company’s moves, if they continue to be successful, could speed up the adoption of AI and automation in banking and the finance industry. [Bold added for emphasis.]
This article made me think about how the banking industry has been changed by automation over the years. As an example, in the summer of 1966 when I was teaching at Michigan State University, I was director of a National Science Foundation Summer Institute for secondary school teachers. We did a field trip to a local bank and got to see a check-sorting machine in action. The paper checks were of many different sizes and shapes, but all contained an account number and bank routing number printed in magnetic ink. The machine could “read” such information at the rate of 2,000 checks per minute. I was very impressed by its speed. While I was watching it did not jam or shred a check.
Of course, the machine could not read the amount of the check. It could only read machine-printed, magnetic ink characters. Contrast that with today’s automatic teller machines (ATMs). I now routinely use an ATM to deposit hand-written checks. The fact that the ATM can “read” the hand-written amount on the check represents an important breakthrough in artificial intelligence.
At the current time, there are about 3.5 million ATMs in use throughout the world (Wikipedia, 2018). Typically, these ATMs are available 24 hours a day, seven days a week, and can dispense cash on request as well as accept checks or cash for deposit. Suppose, for example, that one ATM does as much work in a 168-hour work week as a human bank teller does during a 40-hour work week. That would mean that roughly 3.5 million tellers have been replaced by ATMs. Of course, some new jobs were created, such as manufacturing, installing, and servicing the ATMs, keeping them loaded with cash, and so on. But, the net result is less jobs.
Next, think about using credit cards. Roughly speaking, a purchase made using a credit card is like writing a check. But, for many credit card users, use of a credit card is also a process of taking out a loan. In the United States, well over a half-billion credit cards are in use (Holmes, 6/22/2016). Quoting from this reference:
American consumers collectively are flirting with $1 trillion in credit card debt in mid-2017, according to the Federal Reserve consumer credit report.
The top four issuers provide more than 57 percent of all the cards issued by 5,231 banks.
There were some 14.5 billion U.S. general purpose credit card transactions in the first six months of 2015, accounting for more than $1.4 trillion in purchase volume. [Bold added for emphasis.]
The number of transactions per year has increased slightly since 2015. Imagine the amount of labor that was saved by the approximately 30 billion credit card transactions in 2017, where no paper checks had to be processed. Again, automation is reducing the number of banking jobs.
Aha, and hmm. As noted above, credit cards are a convenient way to borrow money. Think about “the good old days” when a person might go to a bank to borrow money. The process involved working with a bank loan officer whose income was at approximately a “middle class” level, someone who could provide advice regarding the loan. The trillion dollars of credit card borrowing mentioned above was done without the use of any bank employees, and without any advice/counseling by a loan officer! The indiscriminate use of credit cards is now a contributing factor in many bankruptcies.
We now take for granted the bank industry use of ATMs, credit cards, and debit cards, and accept the resulting decrease in banking jobs. This technology is rather low-tech relative to many of the newer uses of Artificial Intelligence that will likely decrease bank employment.
The Swedish bank mentioned at the beginning of this IAE Blog entry believes automation will decrease its workforce by 50% over the next ten years. Currently, there are about 1.5 million bank employees in the United States (Statista, 2017). Suppose that, through a combination of automation and competition, such a 50% reduction occurs at banks throughout the United States over the next ten years. This would mean the loss of about 750,000 jobs that include quite decent benefits.
You realize, of course, that the cost of providing employee benefits is a major expense to an employer. In the United States, we are seeing a significant trend toward creating jobs that provide less than 30 hours per week of employment, and these jobs do not require the payment of employee benefits. As automation facilitates changes in job descriptions and hiring practices, an increasing number of job openings are in positions of less than 30 hours per week. Quoting from (Zenefits, n.d.):
Requirements around employee benefits are first based on the number of full-time employees a company has. If the company is an Applicable Large Employer (ALE), an hourly employee becomes eligible for benefits if the number of hours they work meets or surpasses full-time work. The Affordable Care Act and the IRS define a full-time employee as one who works at least 30 hours a week or 130 hours a month on average. Employees who will be working full-time should be offered benefits based on the company's Waiting Period.
What You Can Do
Be aware of how automation is changing the entire world of commerce, not just the world of banking. Assess your own education to be sure that you are prepared for this changing world. Perhaps of even greater importance, do what you can to make sure that all of our children are getting an education that helps to prepare them for these ongoing, rapid changes. For a discussion of ideas on improving education through more effective uses of Information and Communication Technology (ICT), see my new free book, The Fourth R (Second Edition) (Moursund, July, 2018.)
References and Resources
Holmes, T. (6/22/2016). Credit card market share statistics. Retrieved 8/4/2018 from https://www.creditcards.com/credit-card-news/market-share-statistics.php.
Indeed (2018). Banker salaries in the United States. Retrieved 8/4/2018 from https://www.indeed.com/salaries/Banker-Salarie.
Moursund, D. (July, 2018). The Fourth R (Second Edition). Eugene, OR: Information Age Education. Read online at http://iae-pedia.org/The_Fourth_R_(Second_Edition). Download the Microsoft Word file from http://i-a-e.org/downloads/free-ebooks-by-dave-moursund/307-the-fourth-r-second-edition.html. Download the PDF file from http://i-a-e.org/downloads/free-ebooks-by-dave-moursund/308-the-fourth-r-second-edition-1.html.
Statista (2017). Leading banks in the United States as of December 31, 2017, by number of employees. Retrieved 8/4/2018 from https://www.statista.com/statistics/250220/ranking-of-united-states-banks-by-number-of-employees-in-2012/.
The Download (7/30/2018). Robots are working for Sweden’s banking industry. MIT Review. Retrieved 8/2/2018 from https://www.technologyreview.com/the-download/611744/robots-are-working-for-swedens-banking-industry/?utm_campaign=owned_social&utm_medium=social&utm_source=twitter.com.
Wikipedia (2018). Automated teller machine. Retrieved 8/1/2018 from https://en.wikipedia.org/wiki/Automated_teller_machine.
Zenefits (n.d.). At what point does an hourly employee become eligible for benefits? Retrieved 8/4/2018 from https://www.zenefits.com/answers/at-what-point-does-an-hourly-employee-become-eligible-for-benefits/.