The coming tsunami of automation

“All for ourselves and nothing for other people, seems, in every age of the world, to have been the vile maxim of the masters of mankind.”

–  Adam Smith, Wealth of Nations, 1776


 

Real wages in the US have been flat for decades while productivity has continued to make substantial gains.

 

As we’ve heard time and time again, the increased profits from increased productivity are going to a small number of executives and shareholders rather than back to workers as increased wages or meaningful job creation in departments like R&D. The distribution of wealth continues to shift further and further right into the hands of just a few thousand Americans.

 

America's wealth inequality

 

As this inequality in income continues to grow, Americans are conflicted on the solutions. Pushed primarily by Democrats, entire states are joining the increasing number of large cities whom are now attempting to implement, and with recent success, an increase in their minimum wage to $15 an hour. Just a few days ago, on April 4th 2016, California’s Governor Jerry Brown signed into law a bill that will increase the state minimum wage to $15 per hour by 2023. Although passed, this $15/hour wage for the entire state has a lot of opposition and many argue makes little sense once you leave urban centers. Economists, even some liberal ones, view a state wide $15/hour wage hike detrimentally high and jobs will have to be cut as a result.

New York took a more sophisticated approach to their minimum wage increase. On the same day as California’s bill passing, New York’s Governor Andrew Cuomo signed into law a geographical, business-size based law a minimum wage increase. In New York City, the minimum wage will rise to $15 an hour by 2018 for businesses with 11 or more people. Those with 10 or fewer employees will see a slower rate rise which will hit $15/hour by 2019. Farther out from NYC wages will rise even slower, reaching $15/hour by 2021 and $12.50/hour for more rural businesses by 2020 with an unknown, but expected increase to $15.

 

With two of the most populated states increasing their minimum wages, who will benefit?

Large and rapid increases in minimum wages will affect many more people than only those making federal or state minimum wages. A vast majority of low wage workers are making above the federal minimum wage (93%) but still less than the proposed $15/hour. Twenty Million people in the US are considered “low wage” earners, making less than $10.10/hour (which is the federal minimum wage Obama has pushed for). Considering tipped workers as well, an estimated 24 million people would be effected by a minimum wage increase to just $10.10/hour, millions more would be effected by a $15 hike.

 

The Pew Research Center has recently updated their minimum wage / near-minimum wage (<$10.10/hour) data-set and shown restaurants and food service is the main industry utilizing minimum/low wages with nearly 4 million near-minimum-wage-employees. Grocery stores are a far second at nearly a million. Of these minimum wage workers, nearly all are tasked with providing simple services: operating cash registers, helping customers and stocking shelves, cooking, waiting tables, and cleaning. At the very least, tens of millions of low-skill, low-wage employees will be seeing a large increase in their wages should $15/hour (or even $10.10) become the new minimum. That is, if people still have jobs.

 

 

The economic effects of increasing the minimum wage remain uncertain. Will businesses need cut jobs to stay open? Or will the increased cost just be passed on to consumers? If so, how much will product prices increase? How much more spending will take place when people have more cash? Or will cost of living simply tick up as much as wages? While uncertain, we do have some estimates and small scale experiments to glean insights into these questions.

The Congressional Budget Office estimates a federal minimum wage raise to $10.10/hour will result in the loss of 500,000 jobs while 16.5 million workers will see a substantial increase in their wages. A study from Purdue University estimates an tiny increase of 4% in food cost for the customer if wages were to increase to $15/hour.

However, this 4% claim is hotly contested. The Heritage foundation found fast-food prices would need to increase 38% at a $15/hour wage. This indeed may be closer to the truth especially for smaller, non-franchise restaurants and businesses. After Oakland California’s minimum wage was rapidly increased to $12.25/hour, some restaurants had to up their prices by 20%. However, in San Jose when the minimum wage increased from $8 to $10 in 2012, UC Berkeley researchers at the Institute for Research on Labor and Employment found food prices rose 1.45% on average. Further, wage increases had no employment effects and wage increases were simply passed onto customers. As always, these results too are contested.

 

 

Forward looking for New York, the same group at UC Berkeley expects 36% of NY’s work force being affected by the $15 minimum wage increase, with the vast majority (~80%) being restaurant staff. They further claim prices will only need to rise 0.14% each year (that isn’t a typo). I suppose we’ll need to wait and see if the data are correct.

Could adjusting the minimum wage to $15/hour be too much? In some states where the cost of living comes no where near that of NYC and LA, $15 may be too much. And indeed there are no plans for the federal minimum wage to reach heights of $15. But could $15 too high for urban centers too? What about the rural areas of California who will see wages rise as fast as Silicon Valley?

A small experiment has been running in Seattle for about a year now, when in April 2015 a city wide gradual wage hike to $15 began. On April 1st minimum wages were increased to $10-$11 (depending on geography) and further increases took place January 2016 to $12-$13. Seattle is no where near NYC or LA in terms of cost of living, but it is still considered to be in the top 10 most expensive cities to live in. This wage increase should be a welcomed economic boost, no?

The short term results are worrisome. Employment is down, unemployment is up, and the jobless rate is up. All this turmoil while the US as a whole continues to see the unemployment rate drop to historic lows. It could be noise and basic volatility or it could be the start of a catastrophic trend. Preliminary price analysis after wages rose to $11/hour has found that in just six months, restaurant prices have increased 7.7% on average, 55 times higher than UC Berkeley estimated would take place over an entire year in NYC. However, it is too early to definitively state if this increase was due to higher wages or other economic factors. I think it is worth pointing out as well, no increase in the price of groceries in Seattle was found for the same period indicating the cost of living may not be too severely impacted by rising wages.

 

 

If minimum wage increases result in prices rising in step with wages as well as job loss, they may not be economically sensible.

Will business owners benefit?

From minimum wage proponents, business owners should see substantial benefits in increased wages. Employees will be happier with higher wages, less stressed about making ends meet, and therefore be more productive and less likely to leave, reducing turnover and increasing customer satisfaction. These employees will have more money to spend and boost the economic vigor of their community. Whether using Pew or Berkeley Data, it seems safe to say approximately 20-30% of the population will have more money to spend should wages increase to at least $10.10/hour. That translates to tens of millions of dollars in urban centers like New York City. Should wage growth outpace the cost of living, businesses will theoretically see more customers coming through the door, more revenue, and ideally stable, if not growing, profits. There are some data backing these claims as well. Also, 600 of the America’s top economists support a federal minimum wage increase to $10.10/hour, stating not only will jobs not be lost, but they will be created and an immediate boost of 0.3% GDP would follow. It seems like higher wages should be good for everyone (barring the Seattle and Heritage Foundation data) if the proponents are correct and we have some data pointing in that direction already.

McDonald’s has recently provided its company owned store employees (~10% of McDonald’s employees, most are franchised operations that are not affected by this wage increase) with a small boost to their wages, from $9.09 to $9.90/hour, close to the proposed $10.10/hour federal minimum wage. It is important to note, at the same time, McDonald’s closed hundreds of underperforming stores as well.

The preliminary reports from McDonald’s look good on the surface; turnover is down, employees are more satisfied and so are customers. Turnover is a huge cost for business owners as they have to spend resources to hire and train new people. If wage of $9.90/hour is enough to keep employees satisfied and customers coming back for more, wage increases may be a huge benefit to business owners. As you can see from the chart below, McDonald’s stock price is finally increasing possibly as a result of these increased wages. And although franchisees are reportedly not too happy about corporate raising their store’s wages, these results will hopefully encouraged them to reconsider the benefits of higher wages.

It is important to keep in mind that a wage increase of ~10% from $9.09 to $9.90 is very different from a ~65% increase from $9.09 to $15; a hike NY and CA McDonald’s franchisees will soon be forced to undertake. It is estimated a $15/hour minimum wage could force business owners to accept a 77% reduction in profits in order to stay in business. A slow season could entirely end a business if profits are cut too low. Only time give us the true results.

Although the outcome of increased minimum wage is unknown, I’d like to be optimistic and believe it is the morally correct path to follow.

However, there are some certainties in economics which may hinder any meaningful change: business growth

Corporations, especially public ones, are expected to grow profits. Quora and Reddit both have good explanations on why growth is so important. Simply, people want their money to appreciate and the stock market offers a vehicle to possibly get a high level of appreciation, historically 7% a year, relative to safer, lower interest methods such as savings accounts, the money market, and bonds. Once their profits slow or stagnate, shareholders will expect growth via dividend payments and stock buybacks. Stock prices are always forward looking. If a company is not expected to grow, stock prices will stagnate or fall.

For example, McDonald’s stock price has been stuck in a rut for the past few years (mid 2011 until mid 2015 when the CEO was replaced and forward looking plans, like closing underperforming stores and increasing wages, were put into place). This stagnation in price is primarily due to the key restaurant financial metric called “same-store sales”. It simply measures all the sales of the same store over a period of time, usually quarter to quarter. If a company is growing, same-store sales should increase every quarter. And McDonald’s same-store sales have been falling for a while (Right graph), explaining the stock price stagnation. Had McDonald’s not paid investors a nice and ever increasing dividend each quarter and spent tens of billions of dollars buying back its own stock (instead of paying employees more), the stock price would have likely fallen, rather than only traded horizontally.

 

MCD McDonald's Corp. monthly Stock Chart

McDonald's same store sales

 

Executive Compensation

Another reason for the importance of growth is executive compensation. Executive compensation is more often than not directly tied to stock performance in the form of stock options. For example, a CEO may be given the right to buy their company’s stock in one years time at $100. If they can successfully lead the company and the stock price goes above $100, to say, $120, they are permitted to buy the stock at $100 and immediately turn around and sell it for the market price of $120, pocketing the $20 difference as profit (multiplied by hundreds to hundreds of thousands to even millions of shares). Should the company do poorly and the stock price fall, their stock options are worthless. Since the late 90’s, executives often make much more off stock options than they do from their salaries and benefits. Furthermore, this incentivizes executives to care about short term, rapid, even made up growth more than long term, slow-but-stable growth of the company because the immediate performance under their guidance is directly linked to their compensation while long term they will likely be off at another company.

Hopefully these examples gives you an idea as to why companies are expected to grow and why executives want them to grow at any cost. However, if costs are about to rise due to higher wages, company profits, and in turn their stock price, and in another turn executive compensation, will fall. So long as companies can continue to grow their profits at a fast enough rate, the increased payroll costs could be absorbed. However, in the short term at least, payroll costs for companies reliant on low-wage workers will likely see a dip in profits. It gets worse though…

Profits across the US are falling and expected to keep falling.

 

Quarter 1 of 2016 looks like it will be even worse at -7.5%. And this of course, in the world of publicly traded companies, cannot happen. Executives have to pay the mortgage on their mansions and fuel their yachts and of course pay their employees. Shareholders expect share appreciation or will jump ship for other companies/sectors/investment vehicles. Perhaps when American’s cannot even afford to live, company profits suffer from decreased spending.

Either way, this clearly needs to be fixed. How can profits be increased? Shareholder dividends absolutely cannot be cut; once this happens the stock plummets as investors fear the worst, such as the company heading to bankruptcy. Stock buybacks could be done away with however this too will surely upset shareholders. That leaves cost cutting and price raising.

Raising prices

For the service industry, especially restaurants, profit margins tend to be small. Recently recovering from the recession restaurant profit margins are back up to about 5%. Fast food profits tend to be even smaller, around 3%. There is not much wiggle room to be mess up your cost/profit ratio even for a short time period. Moreover, it is restaurants and other low margin businesses, like grocery stores (1-3% profit margin) and Walmart (3.53% last quarter), who rely on minimum wage workers. If their entire workforce has a wage increase, a company could end up going under. I’ll continue to use restaurants as my example but will expand on automation in many more service industries.

Raising prices is common choice to raise profits. However, price comes down to what the customer is willing to pay for a product, not what a business needs to turn a profit. McDonald’s could not start charging $15 for a BigMac. For fast-food, a 1% price increase leads to a 0.95% decrease in sales on average.

 

People instead go to lower priced competitors or, for the same price, go to a restaurant offering higher quality food. Indeed this seems to be the case. People are opting for fast casual restaurants like Chipotle and Panera Bread, or inexpensive restaurants, like IHOP or Chilli’s, for better food for not that much more cash. And McDonald’s is already raising its prices: Big Macs continue to cost more and more each year while the dollar menu becomes sparse.

I do not mean to entirely single out McDonald’s, they are just a big company with a lot of data on them. All companies, especially in the food service industry, have very small margins and fierce competition. If prices get too high too fast, customers will take their money elsewhere, decreasing sales and hurting profits.

 

 

Mandatory, city/state wide increases in wages do offer some ammunition against price increase issues. All businesses will be hit with the wage increase at the same time and could, theoretically at least, all raise their prices together (LOL). If McDonald’s indeed need only raise the price of its food 4% (like Purdue claims) while wages are raised ~50% from $9.90 to $15, people will have plenty of money to go to McDonald’s (or other, higher quality restaurants, or pay off their student loans, or try to buy a house). However, if that price increase needs to be more like the Heritage Foundation 38%, a 50% wage increase will be completely meaningless to the low wage worker as the price of living simply upticks in step with their wages (if they still have a job). However, with rising wages, prices will have to increase by some amount and we will definitively find out just how much over the next few years in California and New York.

Cutting costs

Along with raising prices, costs can be cut. Cutting down on fixed costs like rent, insurance, and utilities can be difficult but of course doable. Food and materials often falls under the fixed cost umbrella as well. Restaurants can be more efficient and reduce food waste as much as possible, but if you have 100 customers ordering fries, you need 100 orders worth of fries.

Payroll is a massive proportion of a company’s costs. For example, on average 32% of a restaurant’s cost is food while 33% is payroll. Your average McDonald’s is a similar story, where annual food costs come in around $810,00 and payroll comes out to be $702,00. Companies will not only need to pay their workers more, but they’ll also be paying the tax-man more in payroll taxes. After cutting hours and freezing hiring, low-wage reliant businesses often have no other option but to cut staff in order to reduce costs (since “salaries” and “bonuses” do not exist at your average McDonald’s store).

McDonald’s of course needs people putting together burgers and taking orders. Grocery stores need shelf-stockers. Almost all places of business need cashiers. Or do they? Cutting down on payroll, since it makes up such a large proportion of cost, can have a large impact on the bottom line. Moreover, if payroll costs are going to inevitably increase with minimum wage laws, cutting payroll costs is looking very enticing. Finally, as technology advances and becomes more and more affordable, automation seems to be an obvious solution to replace costly humans.

Simple tasks are simple to automate

The Cashier

A cashier’s job is to enter in products and carry out a financial transaction. Both of these are very simple tasks and easy to automate.

People are becoming more accustomed to using electronics in their everyday lives. Apple has sold nearly 300 Million iPads since 2010. The age kids begin to use electronics seems to decline every year; results from a preliminary study found children as young as 1 year old know their way around an iPad. Pediatricians even recommend toddlers use iPads for 30-60 minutes a day. On the other end of life, older people make up a large portion of iPad users. Go into any Apple store on a given day and retirees are abundant (this anecdote may be a bit biased seeing as my data point is from South Florida, retirement capital of the US).

 

Emarketer

 

The point of all this is technology is exceedingly easy to use and people of all ages are becoming comfortable using it, as opposed to just a decade ago when iPads didn’t even exist. Rather than have an employee use a computer to enter in an order or scan a product, why not have the customer use the computer for the same tasks for free? Or no scanning at all? In the past few years self checkouts have become abundant in retail and grocery stores. My local CVS in Boston installed three self-checkouts with one person manning them all. The endless rows of empty checkout lanes in Walmart are now being replaced with dozens of self-checkout counters.

Walmart self checkout minimum wage
Walmart Self-Checkout – Photo Credit: Beth A. Birmingham
CVS Self-Checkout – Photo Credit: http://www.chaindrugreview.com/

 

As people get more comfortable with technology, such as using iPads in their every day lives, I think people will become more accepting of retail self-checkout especially as the systems are refined (cue “unexpected item in the bagging area”) and methods to combat theft are worked out.

Self-checkouts are not limited to retail cashiers however. Self-order/self-checkout kiosks are becoming more and more popular in the food industry. Under the guise of the “Create Your Taste” campaign, McDonald’s has been rolling out self-order kiosks where you get to create your own receipt. Large casual dining chains, like TGI-Fridays, Applebees, and Olive Garden are implementing order taking, bill paying tablets at their tables.

 

McDonald's Self-Order Kiosk minimum wage
McDonald’s Self-Order Kiosk Source: Me
Applebee's table tablet minimum wage
Applebee’s table tablet

By replacing cashiers, hotel check ins, and wait staff with kiosks and tablets, payroll costs could be significantly reduced. A single waiter could wait many more tables if all they needed to do was bring out food and top up drinks. And while McDonald’s claims their kiosks will not effect the number of employees at their stores, analysts and franchisees are being more realistic: this is a clear move to automation in order to combat increasing minimum wages.

Side note: Computers are incredibly cheap

Computing continues to get cheaper with the price of a million transistors down to just 5 cents. Small but powerful computers like smart phones and tablets, continue to get cheaper too as they become popular rather than a niche product for electronics nerds. Because of this, implementing tablets, kiosks, and self-checkouts is becoming exceedingly economical. McDonald’s kiosks are expected to cost franchisees $120,00 – 160,000, approximately 25% of payroll costs for a single year. And if we’re being realistic and McDonald’s cuts staff at these locations, the kiosk’s costs are quite small in the long term (especially if a drive-through version is developed). With self-order/self-checkout tech, will fast-food chains need front-of-store staff at all in the future? I would expect they will not as the trail is already being blazed…

 

 

Easta has is a fast casual restaurant with no front of store staff at all. Walking in there are only a wall of tablets ready to take your order and money. Once you place your order, cooks in the back prepare your meal and place it in your designated cubby. It isn’t hard to imagine every fast food chain following this model as our population grows more comfortable with electronics and self-service.

 

Eatsa - San Francisco
eatsa via http://www.contemporist.com/

 

Moving forward, I think it is safe to say at the very least self-order and self-checkout tablets and kiosks will become the norm for the food industry and many retailers such as corner and big-box stores. As minimum wage hikes are implemented, especially to $15 an hour in some cities and states, it seems like a no-brainer to lighten up on staff anywhere you can. In a just world, these employees would just be reassigned to clean up or food prep. However, seeing how 1.5 Million cashiers are low-wage workers and may soon be making $15 an hour, I imagine many employees will simply be let go.

Food Prep

A basic algorithm to follow

Big Mac: 1) Toast buns 2) Get Big Mac box 3) Buns in box 4) Add sauce 5) Add toppings 6) Add cheese 7) Add patty 8) Assemble – Done.

Computers are exceptionally good at executing a series of commands step by step. Implementing software to translate orders to prep instructions would be trivial. Computers also don’t make mistakes; if a computer gets a “no pickles” instruction it won’t accidentally add them. Computers also don’t need healthcare or breaks during the work day or vacations. So why aren’t automated systems making our burgers?

I’m not entirely sure. I would assume it comes back to costs. However, since minimum wages are on the rise and the cost of computing is coming down, we may see food prepping robots being rolled out. Hell, we already trust robots to put together our cars with sub-millimeter welding and riveting accuracy; throwing some ingredients together cannot be all that difficult for our robot overlords.

 

 

Indeed it’s not. With just a $40 Arduino Microcontroller (minicomputer), two UK electronics enthusiasts, Ian Cooper and Jake Osborne, have created an automatic cocktail machine dubbed “The Inebriator” (below). They have of course put a lot of time, effort, and clever ingenuity into building and programming it, but for a large public company the R&D costs would be unsubstantial. Imagine an automated barista at Starbucks. Throw your order into their smartphone app, set a pickup time, and whala, Barista 3000 has your triple shot venti mocha soy latte ready for you at 3:05pm.

 

 

After watching the Inebriator, imagine a similar burger (or any food) making robot. A conveyor belt continuously moves as orders come in from table side tablets. Once an order comes in, a bun is released into a conveyor toaster which lands on the main conveyor belt. Simple pumps dispense sauces. Toppings fall through chutes as the burger passes by. A spatula-bot moves a burger hot off the grill onto the lettuce bed. A fresh slice of cheese is cut and falls onto the burger as it rolls by. The bun top falls out of its toaster and onto the melting cheese. A simple lever closes the box’s lid as the burger falls down a chute. Over an intercom GLaDOS announces, “Order 42 is ready to be collected”.

I’d be extremely surprised if fast food chains were not working on such a robot. However, even if they’re not, a group of entrepreneurs at Momentum Machines are. Their burger machine can turn out 400 burgers an hour and they estimate it can save fast food restaurants $135,000 in wages every year, about the same as self-order kiosk’s one-off cost… Want to invest in Momentum Machines and make millions? Well you’re too dumb for that.

Momentum Machines' Burger-Bot Minimum wage killer
Momentum Machines’ Burger-Bot
Fresh-Sliced Tomato
Fresh-Sliced Tomato in Momentum Machine’s Burger-bot

 

I imagine the McDonald’s, and all fast-food restaurants for that matter, of the not so distant future needing just 1-2 employees to take deliveries, stock equipment, and clean up. Basic automation is already well entrenched for places like Krispy Kreme donuts who have raw dough entering one side of a machine and finished donuts coming out the other. As these meal-prep robots are developed and adopted more, and computing costs continue to fall as wages rise, I could even see fast casual chains like IHOP and Chilli’s adopting similar robots to replace at least some of their cooks in making simple food items like french fries, eggs, pancakes, and of course burgers and sandwiches.

 

Inventory Management – Shelf-Stockers/Auditors

The once universal teenage job is now the way of life of many American adults, but are robots coming to a grocery store near you?

Amazon’s warehouses look like a miniaturized city: robots crawl around in perfect harmony, bringing packages from one place to another. All humans are left to do is grab, scan, and pack your new widget. These bots stand to save Amazon upwards of a Billion dollars. Can robots handle inventory outside of warehouses?

Simple, mobile, people friendly, one armed UBR-1 robots were slated to come onto the scene of warehouses, factories, and possibly grocery stores. The robots were designed to work with and around people. While not explicitly built for shelf-stacking, it is easy to see how UBR-1 could stack shelves, unload food stuffs, and organize products behind the scenes of grocery stores similar to Amazon bots. (Note: Unbounded Robotics, makers of UBR-1, seem to have gone out of business for unknown reasons)

 

Amazon Kiva
Amazon warehouse robots

 

UBR-1 robot stacking
UBR-1 Robot

A more specialized, shelf-auditing robot is in the works from Simble Robotics. Simble’s “Tally” automatically audits the entire store to check which products need to be re-stocked, reducing labor time of human employees. I imagine Tally and an UBR-1 clone being great friends working together. Or better yet, having an all-in-one shelf-auditing, shelf-stacking robot baby.

 

Even with the low cost of computing, these robots, for now, are expensive. UBR-1 was expected to cost $35,000. Tally’s price has not been revealed. On the flip-side, Pew Research found grocery stores rely on nearly 900,000 low-wage workers. If their wages increase substantially, robots which don’t need a wage, payroll tax, or health benefits may become cost efficient. Further, if these types of robots are adopted by grocery stores, retail stores such as Walmart and Best Buy may follow suit.

No Stores at all?

On top of payroll, other costs like rent, utilities, and insurance are a huge for many restaurants and retailers; why not get rid of all the costs?

Vending machines are abundant in Japan with one for every 23 people. Just as RedBox (and Netflix) killed the video rental store, will vending machines kill brick-and-mortar stores? Grocery stores may be slow to change, especially for items like fresh meat and produce. However, retail stores may have something to worry about. In Japan you can easily get all your essentials (and more) from vending machines: rice, bread, eggs, noodles, pizzas, wine, pet food, clothing, shoes, bike parts, you get the picture. Many airports in America are rolling out high-tech product vending machines. Best Buy vending machines are quite common selling headphones, chargers, and even iPods. I would expect Americans as a whole will take a lot of warming up to reach Japan level vending machine culture, but if wages are rising and property prices continue to skyrocket, the already developed tech could be a low cost alternative to retail outlets and their massive overhead costs.

 

 

Janitors and building cleaners

As we go down Pew’s list, we come to janitors as the final occupation to automate. Automation of janitorial jobs seems just as inevitable as the rest of the aforementioned jobs. Roomba vacuums start at just $375 bucks now a days. More industrial floor cleaning robots have already been developed and are hitting the market. Taski by Inellibot Robotics offers industrial vacuum, scrubber, and sweeper/scrubber combo robots. Robot cleaning businesses using these bots are beginning to pop up around the US. As our usual conclusion goes, as minimum wages rise and electronics continue to get cheaper, AI filled janitor bots may replace at least some expensive humans on cleaning staffs.

Intellibot Cleaners
Intellibot Cleaners

 

Even more automation

I thought it would be important to cover and spend the most time on the majority of jobs that rely on near-minimum-wage workers. There are of course many other service jobs, and non-service jobs, that can be automated which I will just quickly touch on as I see I’m already hitting 4,000 words with this sentence. The final one on the list which I will not cover in detail is retail salespeople (thought they often work as cashiers too). Assuming shelf-stocking robots can handle clothes and electronics, salespeople to answer questions would still be useful to many retail stores. However, Best Buy is beginning to cover that too, with Chloe their customer service robot who can fetch products for you. Throw a store-specific Siri-like AI onto a shelf-stocking robot and customer service is handled for the majority of stores.

Below are a few more jobs, including high paying ones, that may go away with robots and artificial intelligence.

The list could go on and on.

 

Conclusions

Repeated below is the Pew Research Group finding on who are minimum wage workers. We covered nearly everything with the exception of schooling specifically, however school cafeteria staff and janitors, those likely to be paid by schools near-minimum-wages, have been covered. So what happens if they all go away? Will California and New York businesses be the trail blazers for automation now that $15/hour minimum wages are inevitable? To give some perspective, the new Rasberry Pi Zero single board computer costs just 20 minutes of a $15 an hour wage. With the cost of computing so ridiculously low, will more start-ups pursue robotic and AI automation to fill the inevitable higher minimum wage market?

 

 

Or will robots just remain in the movies? Robots have been threatening to take jobs for decades. Perhaps everyone will just come out ahead as many economists and people hope a minimum wage increase will do. Employees making up 20-30% of the population will have more money to spend, their lives will be improved, businesses will grow, and this despicable wealth inequality thing will begin to be tackled. Surely the Walton’s are content with their current $144 Billion and can afford to experiment with paying workers a morally acceptable wage instead of relying on tax payers to subsidize their employees to the tune of billions of dollars a year?

Well I sure do hope so. I hope Seattle is just in a weird hiring phase and employment comes back up. I hope NY’s and CA’s economies boom as millions are lifted out of poverty and many cities and states follow their example.

On a similar note that deserves attention, historically, technology does displace jobs but new jobs are invented. There was no such things a software engineer in 1910. App developers didn’t exists in 1990. The majority of experts surveyed by Pew said that by 2025, robots and AI will probably not have a negative impact on employment for humans. That majority however was only 52%; 48% thought robots and AI would have a negative impact and displace more jobs than they create by 2025.

Because this time it may be different. Jobs may not be made at the rate they are lost. Robots are cheap. Software and AI improves every day. And people are just so expensive, they always have been in business. Aside from needing more high-skill engineers and programmers, I think that many, if not all, low-skill jobs will disappear with no skill-level equivalent over the next few decades in America.

 

Future of AI/robotics

 

And although I want to remain optimistic, I must admit there is a sizable portion of my brain that is reminding me of the other force acting against employment for humans: Just how selfish human beings are. Adam Smith pointed out in Wealth Of Nations (1776) that the rich controlled the government and its policies. Throughout all of human civilization, this was the case. And today is no different; American policy is shaped by the selfish elite. If they can make more money by automation, they will do so. Perhaps now is the time they allow minimum wages to increase because they know the automation technology is here to replace workers.

 

Distribution of income growth: Trickle Down Economics is simply not true
Distribution of income growth: Trickle Down Economics is simply not true via Pavlina R. Tcherneva

 

The 2007/8 financial crisis is just the most recent example of wide spread selfishness where short term gains are all that matter despite the massive and disastrous costs. And they got away with it. They got bailed out by you and me and it’s only a matter of time before it happens again.

An even more recent event, though the full repercussions remain unknown, is the Panama Papers: showing the very people running governments may have secret financial stakes behind their legislative and political motivations.

I find the selfish drive that we may have inherited from our genes could be entirely detrimental to our prosperity, perhaps even our continued existence, as a species. Examples of the devastating suffering inflicted by the selfish nature of a few powerful men onto the many can be seen in every epoch of human history. Just look how selfish anthropomorphic climate change ‘opponents’ are; the livelihood of future generations have been knowingly destroyed for corporate profits, offshore account balances, and political power.

Indeed, these selfish, profit driven institutions are already planning for the robotic future and how they can maximize their profits from it. Bank of America’s Merrill Lynch, their investment branch, has recently put together a 300 page report on the “Robot Revolution” which includes AI. They estimate 47% of the workforce is up for automation by robots and AI and that labor costs could be cut by ~$9 Trillion dollars in 10 years. On the very next bullet point, they elucidate where clients should invest to maximize their investment returns based on this information.

The Economist Magazine has recently reported that top tech firms are poaching academic AI and robotics researchers from top universities. The big companies are positioning themselves ahead of the coming demand for automation to have the solutions needed to replace costly human workers.

Finally, manufacturing. Manufacturing jobs in the US which were not lost to automation have largely gone overseas and over-borders. However, overseas manufacturing is beginning to lose its profitability and if wages at home continue to rise, well, something needs to keep profits growing. Chinese workers wages have even risen high enough that iPhones will soon be made by robots too. Not only are services up for automation, but globally abundant manufacturing jobs may never come back as robotics and AI improves.

The warning sirens are beginning to hum. The tech tsunami is rolling in. How high the wall of water will get is still a mystery. But it is coming, and fast. And it’s impact will be felt far and wide by the many at the mercy of the selfish elite.

A solution?

This post had ended up being another failed attempt at being succinct so I might as well keep going. I also find it infuriating when observations of problems are stated by an author and no solutions ever posed.

Unconditional basic income: A form of social security in which every resident of a country is paid a basic income from the government regardless of their financial state. The rich, the poor, the middle (whatever that means anymore); everyone is paid. The idea is, in a capitalist system, taxes are used to fund some level of basic income for all adults. Thomas Paine proposed the concept in Agrarian Justice in 1797. In most circumstances it is not meant to act as a primary source of income but rather as a financial boost to all citizens. A few hundred extra dollars a month to help with rent and food. Should mass unemployment by automation occur, a living basic income may be required.

Pilot programs have been undertaken in smaller communities around the world. Many pilots were not basic income per se but negative tax rates. The city of Utrecht in the Netherlands is beginning to test out conditional basic income of £660 a month with groups of people already claiming benefits. While not a true ‘unconditional’ basic income, they are testing the waters.

Opponents are always quick to point out funding such a system would be difficult and productivity would decrease as people would have less incentive to work. Also, handing out money just seems wrong in capitalism.

However, humans often behave in ways that are not always expected. It may be with a basic income, people are free to pursue professions they are passionate about and productivity will instead increase. Rather than sitting around watching TV all day living off the man, people may opt instead to take the risk of pursuing professions they feel passionate about. We can also drop the welfare abuse case, it just is not a thing that happens often. A minuscule 1.9% of welfare spending was found to be fraudulent or abusive. Also, if AI and robots do displace many millions of workers, be it in 2025 or 2050 or 2100, it is not like many people would have jobs to do anyways.

Funding such a program is indeed a large obstacle though. Proponents point out benefits are already paid out in complex welfare systems in every developed country. By simplifying the system and providing a simple, fixed amount basic income, the overwhelming cost of complex welfare bureaucracies can be avoided and aid in funding basic income. Furthermore, healthcare costs may fall as people with higher incomes tend to be healthier. To compliment a basic income system, America may even become sensible and adopt a single payer health care system like a developed society to further save on healthcare costs.

If the rich actually paid their taxes in the US, rather than hoarding them overseas, America may even be able to afford to provide its people with a basic income. Tens of trillions of dollars are untaxed and hidden away by America’s elite. Corporations skip out on hundreds of billions of dollars in taxes every year. The Pentagon is missing trillions of dollars. Should these systems be fixed, a lot of tax money would be coming back into the US and provide a significant boost in income to the 46.7 Million Americans living in poverty and a substantial financial cushions to millions more. A basic income of $2,000 a year, equating to a substantial 17% boost in income to households in poverty, would cost just $650 Billion dollars. Conveniently, that amount is just about how much companies fail to pay in taxes every year.

Either way, countries may soon need to figure out how to fund a basic income system out of necessity if inequality trends continue, especially if they are made worse by robots and AI. As if people are not mad enough about Wall Street being in bed with Capitol Hill and losing jobs overseas, losing even more jobs to robots and AI may be too much to bear. If millions of Americans lose their jobs and little remaining wealth to robots and selfish greed, the 2nd Amendment may begin to frighten those on Wall Street and Capitol Hill.


Some automation updates since publication:


Pssst, you can watch the full documentary here or on Netflix.


References:

Adam Smith. The Wealth of Nations. W. Strahan and T. Cadell, London. 1776.

Drew Desilver. For most workers, real wages have barely budged for decades. Pew Research Group. 9 October 2014.

Josh Bivens and Lawrence Mishel. Understanding the Historic Divergence Between Productivity and a Typical Worker’s Pay. Economic Policy Institute. 2 September 2015

Carroll Doherty, Juliana Menasce Horowitz, and Michael DimockMost See Inequality Growing, but Partisans Differ over SolutionsPew Research Group. 23 January 2015.

Melissa S. Kearney and Benjamin H. Harris. The “Ripple Effect” of a Minimum Wage Increase on American Workers. The Hamilton Project. 10 January 2014.

Drew Desilver. Where near-minimum-wage workers work, and how much they make. Pew Research Group. Pew Research Group.17 November 2014.

Drew Desilver and Charles Schwarzer. Making more than minimum wage, but less than $10.10 an hour. Pew Research Group. 5 November 2014.

Drew Desilver. 5 facts about the minimum wage. Pew Research Group. 23 July 2015.

David Card and Alan B Krueger. Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania. The American Economic Review, 84(4) 772 – 793. September 1994.

The Effects of a Minimum-Wage Increase on Employment and Family Income. Congress of the United States Congressional Budget Office. February 2014.

Jing Ma and Richard GhiselliThe minimum wage, a competitive wage, and the price of a burger: Can competitive wages be offered in limited-service restaurants? Journal of Foodservice Business Research,

19(2) 131-146, 6 April 2016. DOI:10.1080/15378020.2016.1159889

James Sherk. Higher Fast-Food Wages: Higher Fast-Food Prices. Heritage Foundation. 4 September 2014.

Sylvia Allegretto and Michael Reich. Are Local Minimum Wages Absorbed by Price Increases? Estimates from Internet-based Restaurant Menus. IRLE Working Paper No. 124-15. December 2015.

Michael Reich, Sylvia Allegretto, Ken Jacobs and Claire Montialoux. The Effects of a $15 Minimum Wage in New York State. Center on Wage and Employment Dynamics. March 2016.

Mark J Perry. New evidence suggests that Seattle’s ‘radical experiment’ might be a model for the rest of the nation not to follow. American Enterprise Institute. 18 February 2016

Heather Hill, Jennifer J. Otten, Emma van Inwegen, and Jacob Vigdor. Early Evidence on the Impact of Seattle’s Minimum Wage Ordinance. University of Washington. January 2016.

John Schmitt. Why Does the Minimum Wage Have No Discernible Effect on Employment? Center for Economic and Policy Research. February 2013

Dimitri A. Christakis. Interactive Media Use at Younger Than the Age of 2 Years Time to Rethink the American Academy of Pediatrics Guideline? JAMA Pediatr. 2014;168(5):399-400. May 2014. DOI:10.1001/jamapediatrics.2013.5081.

Aaron Smith and Janna Anderson. AI, Robotics, and the Future of Jobs. Pew Research Center. 6 August 2014.

Thomas Paine. Agrarian JusticeJanuary 1797

CDC Health Disparities and Inequalities Report — United States, 2013. CDC22 November 2013.

James S.Henry. The price of offshore revisited. Tax Justice Network. July 2012.

Will Kimball and Robert E. Scott. China Trade, Outsourcing and Jobs. Economic Policy Institute. 11 December 2014.

 


Creative Commons License
This work by Blake Porter is licensed under a Creative Commons Attribution-Non Commercial-ShareAlike 4.0 International License


If you enjoyed this article and found it informative, consider donating! Funds first go to keeping the website running (domain name/hosting) and secondly go to me. Donations are secure and handled by PayPal.




4 Responses

  1. Prashant Kotak

    “…people may opt instead to take the risk of pursuing professions they feel passionate about..”

    Yes, but which professions? Around 50% of all professions globally are automatable within a decade, and likely over 85+% with two decades.

    By the way, I came across your blog because I often use the phrase “automation-driven unemployment tsunami” and was googling for it looking for another post. I am an (old) coder working in IT. I bang on to people about the sheer speed with which automation is going to unfold (to the point of boredom), but most people, even in IT, simply don’t get it – they look at the low unemployment rate and think I am talking complete nonsense when I tell them things will be very different in as little as 5 years. They also don’t get the point that the reason we have not had a “traditional” global recovery since the 2008 crash is mostly due to the pernicious pressure that software and automation has put on earnings. Most also don’t have any idea of the societal disruption automation will cause and the threat it poses to incomes, lifestyles, and any kind of long term stability, where every bit of know-how people have built up over decades is rendered useless withing half a decade. When discussing this topic, I often hear people say to the effect: “we will come up with new goods and services that machines aren’t able to do create, so people will have work”. To which I answer: but automation isn’t three decades away, it’s not even three years away, it’s here now. Unless you can point, right now, to what those goods and services are, I venture to suggest you have lost the argument. At which point people find the topic both frightening and tedious and switch the topic of conversation to the football game last night.

    Let me just clarify, I am not a Luddite. I know technology-driven change is unstoppable – the chances of even slowing it down are about as good as King Canute holding back the sea. But I am now extremely pessimistic about what is coming. Part of my alarm stems from the speed and simultaneity with which things will happen. A tipping point will be hit sometime within the decade, when lots professions start losing human workers *all at the same time* globally. As there won’t be many professions left people can run to, everyone with any gumption or spirit, everyone who doesn’t want to be “..sitting around watching TV..”, will attempt to retrain into professions that are still viable. The world is full of smart people. That means (to give just an inkling), a lot of taxi-drivers will attempt to learn programming. A lot of lawyers will retrain as Neuroscientists. Imagine this on a global scale, affecting hundreds of millions (perhaps billions) of people. Another concern is about observed anthropological phenomena in peoples who are technologically defeated, that large sections of humanity sink into a kind of stupor, of drink and drugs and depression – think American Indians in reservations or Australian Aborigines.

    As you have analysed the impact of automation so eloquently, I have no doubt that you are aware that technology-driven change won’t stop with automation. Within two to four decades Machine Intelligence will surpass humans. Don’t get me wrong – I’m not talking about sentient MI, the mechanisms that create sentience are not understood (if indeed those processes are mechanistic).
    I thought for over three decades that sentience was the bulwark against MI exceeding humans. I now no longer think that (even though I wish it) because it is becoming increasingly clear that sentience is not necessary for MI to match human-level processing.

    Parking the long term issue that MI, sentient or not),poses to humanity (existential threat), I can see upcoming a period of about two decades, where about a tenth of humanity utilise MI to cause an absolute explosion of creative output in every conceivable field (and in many fields not yet conceived). That will improve both humans and the MI that they create. But humans are stuck in biological timescales – and MI operates in electronic ones, many orders of magnitude faster.

    So at the end of that, MI on it’s own will be making better decisions in any field without humans – human input will add nothing – what then for humans?

    (Writing this post has given me the thought that I should perhaps blog about all this, because I have a lot of thoughts and ideas about this and associated issues).

    • Blake Porter

      “…people may opt instead to take the risk of pursuing professions they feel passionate about..”

      This quote assumed a universal basic income. It meant to offer a solution of what people will do on a universal basic income. Just cause robots could handle all of our agriculture won’t mean people who like to garden will stop gardening. Same with woodworking, programming, etc etc. I suppose my use of “profession” was incorrect and I meant more of, following a hobby or passion with all a person’s time and not having to derive an income from it.

      Yeah I thought quite a bit about the title and though tsunami was a good analogy. We can see this coming now, though out in the ocean the wave may not look that large. However, once it reaches shore it will have a profound impact.

      I’m not even sure people will have the time to really retrain, by the time people realize they’re situation, are driven enough to retrain and not just give up or remain stubborn (see Taxi drivers fighting Uber), apply to programs, do the schooling, find work (if they can), automation will swallow that field quick enough. Let alone people who will even have the funds to retrain.

      I agree with you on the MI front, I’m not even sure we will need to wait for general MI for profound impacts to take over. Just highly specific but highly capable MI platforms would be enough to fully disrupt the modern workforce.

      What then? Who knows. I’m sure we can be optimistic and hope for a world, sort of what I was talking about at the top of this, that people will be free to pursue whatever activities bring them joy. But of course a dystopia of depression and substance abuse seems equally likely. I’d assume, as it seems to be the case for many things, the average will fall somewhere in the middle on a spectrum.

      If you do blog about it let me know! I’d be interested to read what you think, esp from someone with a longer term IT background.

Leave a Reply to Blake Porter Cancel reply