It’s been a while since I last blogged about labour markets and, in particular, about the effect of automation on wages and employment. My most recent post on this topic was on the 14th of April 2018 and it was mostly a reflection on some interesting findings that had been reported by Acemoglu et al (2017). More specifically, Acemoglu and Restrepo (2017) developed a theoretical framework to evaluate the effect of AI on employment and wages. They concluded that the effect was negative and potentially sizeable (for a more detailed discussion see my blog).
Using a model in which robots compete against human labor in the production of different tasks, we show that robots may reduce employment and wages … According to our estimates, one more robot per thousand workers reduces the employment to population ratio by about 0.18–0.34 percentage points and wages by 0.25–0.5 percent.
Since then, I have seen a constant stream of news on my news feed about the development of ever more advanced industrial robots and artificial intelligence. And this was not because of some spooky coincidence (or worse). It has been merely a reflection of the speed at which technology has been progressing in this field.
There are now robots that can run, jump, hold conversations with humans, do gymnastics (and even sweat for it!) and more. It is really impressive how fast change has been happening recently in this field – and, unsurprisingly, it has stimulated the interest of labour economists!
A paper that has recently come to my attention on this subject is by Graetz and Michaels (2018). The authors put together a panel dataset on robot adoption within seventeen countries from 1993 to 2007 and use advanced econometric techniques to evaluate the effect of these technologies on employment and productivity growth. Their analysis focuses exclusively on developed economies (due to data limitations, as they explain) – but their results are nevertheless intriguing:
We study here for the first time the relationship between industrial robots and economic outcomes across much of the developed world. Using a panel of industries in seventeen countries from 1993 to 2007, we find that increased use of industrial robots is associated with increases in labor productivity. We find that the contribution of increased use of robots to productivity growth is substantial and calculate using conservative estimates that it comes to 0.36 percentage points, accounting for 15% of the aggregate economy-wide productivity growth.
The pattern that we document is robust to including various controls for country trends and changes in the composition of labor and other capital inputs. We also find that robot densification is associated with increases in both total factor productivity and wages, and reductions in output prices. We find no significant relationship between the increased use of industrial robots and overall employment, although we find that robots may be reducing the employment of low-skilled workers.
This is very positive news for most – except, of course, for low-skilled workers. Indeed, like Acemoglu and Restrepo (2017) and many others, this study shows that the effect of automation on employment and labour market outcomes is unlikely to be uniform across all types of workers. Low-skilled workers are found again to be likely to lose out and be significantly displaced by these technologies.
And if you are wondering which sectors are likely to be disrupted most/first by automation, the rankings developed by McKinsey and Company (see chart below) would give you an idea of where the disruption is likely to start. Unsurprisingly, the sectors that seem to be the most vulnerable, are the ones that use the highest share of low-skilled labour.
- “The effect of automation on wages and employment is likely to be positive overall”. Discuss.
- Using examples and anecdotal evidence, do you agree with these findings?
- Using Google Scholar, put together a list of 5 recent (i.e. 2015 or later) articles and working papers on labour markets and automation. Compare and discuss their findings.
There have been two significant changes in prices for travel in Bristol. At the end of April, the toll on Brunel’s iconic Clifton Suspension Bridge doubled from 50p to £1 for a single crossing by car. The bridge over the Avon Gorge links North Somerset with the Clifton area of Bristol and is a major access route to the north west of the city. Avoiding the bridge could add around 2 miles or 8 minutes to a journey from North Somerset to Clifton.
The justification given by the Clifton Suspension Bridge Trust for the increase was that extra revenue was needed for maintenance and repair. As Trust Chairman Chris Booy said, ‘The higher toll will enable the Trust to continue its £9 million 10-year vital repair and maintenance programme which aims to secure the bridge’s long-term future as a key traffic route, one of Bristol’s major tourist destinations and the icon of the city’.
The other price change has been downwards. In November 2013, the First Group cut bus fares in Bristol and surrounding areas. Single fares for up to three miles were cut from £2.90 to £1.50; 30% discounts were introduced for those aged 16 to 21; half-price tickets were introduced for children from 5 to 15; and the two fare zones for £4 and £6 day tickets were substantially increased in size.
First hoped that the anticipated increase in passengers would lead to an increase in revenue. Evidence so far is that passenger numbers have increased, with journeys rising by some 15%. Part of this is due to other factors, such as extra bus services, new buses, free wifi and refurbished bus stops with larger shelters and seats. But the company attributes a 9% rise in passengers to the fare reductions. As far as revenue is concerned, indications from the company are that, after an initial fall, revenue has risen back to levels earned before the fare reduction.
What are the longer-term implications for revenue and profit of these two decisions? This depends on the price elasticity of demand and on changes in costs. Read the articles and then consider the implications by having a go at answering the questions.
Clifton Suspension Bridge toll to rise from 50p to £1 BBC News (9/4/14)
Regular Users of Clifton Suspension Bridge will be Protected from the Increase in the Bridge Toll Clifton Suspension Bridge (9/4/14)
Clifton Suspension Bridge Review Decision Letter Department of Transport (24/3/14)
Clifton Suspension Bridge Trust: bridge toll review inspector’s report Department of Transport (8/4/14)
Clifton Suspension Bridge Toll Increase – Account of the May 2013 Public Inquiry The National Alliance Against Tolls (NAAT)
First Bus Bristol fare cuts sees passenger growth BBC News (6/6/14)
First gamble over cheaper bus fares pays off as passengers increase in Bristol The Bristol Post (6/6/14)
Bristol bus fares deal to extend to South Gloucestershire and North Somerset The Bristol Post, Gavin Thompson (12/6/14)
- What assumptions is the Clifton Suspension Bridge Trust making about the price elasticity of demand for bridge crossings?
- What determines the price elasticity for bridge crossings in general? Why is this likely to differ from one bridge to another?
- How is the long-term price elasticity of demand likely to differ from the short-term elasticity for Clifton Suspension Bridge crossings and what implications will this have for revenues, costs and profit?
- How is the price elasticity of demand for the bridge likely to vary from one user to another?
- How is offering substantial price reductions for multiple-crossing cards likely to affect revenue?
- What determines the price elasticity of demand for bus travel?
- What could a local council do to encourage people to use buses?
- How is the long-term price elasticity of demand for bus travel likely to differ from the short-term elasticity?
- In the long run, is First likely to see profits increase from its fare reduction policy? Explain what will determine this likelihood.
Facebook has announced that it’s purchasing the messaging company WhatsApp. It is paying $19 billion in cash and shares, a sum that dwarfs other acquisitions of start-up companies in the app market. But what are the reasons for the acquisition and how will it affect users?
WhatsApp was founded less than five years ago and has seen massive growth and now has some 450 million active users, 70% of whom use it daily. This compares with Twitter’s 240 million users. An average of one million new users are signing up to WhatsApp each day. As the Wall Street Journal article, linked below, states:
Even by the get-big-fast standards of Silicon Valley, WhatsApp’s story is remarkable. The company, founded in 2009 by Ukrainian Jan Koum and American Brian Acton, reached 450 million users faster than any company in history, wrote Jim Goetz, a partner at investor Sequoia Capital.
Facebook had fewer than 150 million users after its fourth year, one third that of WhatsApp in the same time period.
Yet, despite its large user base, WhatsApp has just 55 employees, including 32 engineers.
For the user, WhatsApp offers a cheap service (free for the first year and just a 99¢ annual fee thereafter). There are no charges for sending or receiving text, pictures and videos. It operates on all mobile systems and carries no ads. It also offers privacy – once sent, messages are deleted from the company’s servers and are thus not available to government and other agencies trying to track people.
With 450 million current active users, this means that revenue next year will not be much in excess of $450 million. Thus it would seem that unless Facebook changes WhatsApp’s charging system or allows advertising (which it says it won’t) or sees massive further growth, there must have been reasons other than simple extra revenue for the acquisition.
Other possible reasons are investigated in the videos and articles below. One is to restrict competition which threatens Facebook’s own share of the messaging market: competition that has seen young people move away from Facebook, which they see is becoming more of a social media platform for families and all generations, not just for the young.
Videos and podcasts
Facebook pays billions for WhatsApp Messenger smartphone service Deutsche Welle, Manuel Özcerkes (19/2/14)
Facebook’s WhatsApp buy no bargain Reuters, Peter Thal Larsen (20/2/14)
Facebook Agrees To Buy WhatsApp For $19bn Sky News, Greg Milam (20/2/14)
Facebook Eliminates Competitor With WhatsApp Bloomberg TV, Om Malik, David Kirkpatrick and Paul Kedrosky (20/2/14)
Why WhatsApp Makes Perfect Sense for Facebook Bloomberg TV, Om Malik, David Kirkpatrick and Paul Kedrosky (20/2/14)
Facebook buying WhatsApp for $19bn BBC News, Mike Butcher (20/2/14)
Is Facebook’s acquisition of WhatsApp a desperate move? CNBC News, Rob Enderle (19/2/14)
Facebook’s $19bn WhatsApp deal ‘unjustifiable’ BBC Today Programme, Larry Magid (20/2/14)
Facebook to buy WhatsApp for $19 billion in deal shocker ReutersGerry Shih and Sarah McBride (20/2/14)
Facebook to Pay $19 Billion for WhatsApp Wall Street Journal, Reed Albergotti, Douglas MacMillan and Evelyn M. Rusli (19/2/14)
Facebook to buy WhatsApp for $19bn The Telegraph, Katherine Rushton (19/2/14)
Facebook buys WhatsApp: Mark Zuckerberg explains why The Telegraph (19/2/14)
WhatsApp deal: for Mark Zuckerberg $19bn is cheap to nullify the threat posed by messaging application The Telegraph, Katherine Rushton (20/2/14)
Why did Facebook buy WhatsApp? TechRadar, Matt Swider (20/2/14)
What is WhatsApp? What has Facebook got for $19bn? The Guardian, Alex Hern (20/2/14)
Facebook to buy messaging app WhatsApp for $19bn BBC News (20/2/14)
WhatsApp – is it worth it? BBC News, Rory Cellan-Jones (20/2/14)
Facebook buys WhatsApp: what the analysts say The Telegraph (19/2/14)
Facebook ‘dead and buried’ as teenagers switch to WhatsApp and Snapchat – because they don’t want mum and dad to see their embarrassing pictures Mail Online (27/12/13)
Facebook and WhatsApp: Getting the messages The Economist (22/2/14)
- Are Facebook and WhatsApp substitutes or complements, or neither?
- What does Facebook stand to gain from the acquisition of WhatsApp? Is the deal a largely defensive one for Facebook?
- Has Facebook paid too much for WhatsApp? What information would help you answer this question?
- Would it be a good idea for Facebook to build in the WhatsApp functionality into the main Facebook platform or would it be better to keep the two products separate by keeping WhatsApp as a self contained company?
- What effects will the acquisition have on competition in the social media and messaging market? Is this good for the user?
- Will the deal attract the attention of Federal competition regulators in the USA? If so, why; if not, why not?
- What are the implications for Google and Twitter?
- Find out and explain what happened to the Facebook share price after the acquisition was announced.
The law of demand tells us that when the price of a good falls, quantity demanded will rise. But, firms want to know much more than this. They need to know by how much quantity demanded will rise – we refer to this as the price elasticity of demand (PED) and we can categorise it as relatively inelastic or elastic, depending on by how much demand changes relative to the change in price. The price elasticity of demand is crucial for a firm to know, as it gives them vital information about the best price to charge and getting the price right is probably the most important element in a successful business. As Warren Buffett said in a meeting with the staff from the Federal Crisis Inquiry Commission:
‘Basically, the single most important decision in evaluating a business is pricing power. You’ve the power to raise prices without losing business to a competitor, and you’ve got a very good business. And if you have to have a prayer session before raising the price by a tenth of a cent, then you got a terrible business.’
The grammar may not be entirely correct, but hopefully you get the gist! Should a firm increase price or reduce it? Whatever action it takes, there will be an effect on demand, total revenue and profit. The key question is: what will be the effect? The answer depends on the PED.
If a firm is selling a product for which there are no close substitutes, we would expect demand to be relatively inelastic. This means that the firm can increase the price it charges without seeing any large fall in quantity. On the other hand, if a firm faces a lot of competition and hence there are many substitutes for a product, then demand becomes much more elastic – any increase in a firm’s price will lead to a proportionately larger decrease in the quantity demanded, as customers will simply switch to a cheaper alternative. The article below looks at the concept of price elasticity of demand and how it is used in practice by competing firms.
The importance of pricing power: PEP, CPB Guru Focus (16/10/11)
Pricing strong for Philip Morris in Q3, but volumes also encouraging; dividend yield attractive MorningStar (7/11/11)
- How do we define price elasticity of demand and what formula can we use to calculate it?
- If a firm faces an PED of –5, is its demand relatively inelastic or elastic and what does it mean about the responsiveness of customer demand to a change in price?
- If a firm faces demand that is (a) relatively inelastic (b) relatively elastic, (c) perfectly elastic (d) perfectly inelastic, what should it do to its price? Explain your answers.
- In the article, ‘The importance of pricing power’, is demand for the ‘Daily Racing Forum’ relatively inelastic or elastic? Explain your answer and what it means in terms of the company’s ability to change price.
- Is demand for cigarettes likely to be inelastic or elastic? Explain your answer. What does this suggest about a firm’s ability to pass on taxation and excise duties to its customers in the form of higher prices?
- Based on the data given in ‘The importance of pricing power’ about the change in demand for Campbell’s Soup and PepsiCo, what conclusions can we reach about PED? How could these firms use this information to set prices and maximise revenue and profit?
- Following a change in supply (due to a factor other than price), when will the impact on equilibrium price be larger than the impact on equilibrium quantity?
This podcast is from BBC Radio 4’s Today Programme. It consists of an interview with James Berresford, chief executive of VisitEngland, and Tracy Corrigan, of the Daily Telegraph on the topic of ‘staycations’ – a term used to refer to people holidaying at home rather than going abroad. Staycations are up, but why is this the case; how much have people switched; and is it really a cheaper option?
More people holidaying in England BBC Today Programme (27/8/09)
See also the following articles:
Unemployment Up In Seaside Resorts Despite Era Of The ‘Staycation’ Fresh Business Thinking (22/8/09)
Unemployment up in seaside resorts despite era of the ‘staycation’ TUC (21/8/09)
Haven Holidays sees rise in caravan sales Times Online (26/8/09)
‘Staycation’ Britons reconsider their holiday plans The National (Abu Dhabi) (28/8/09)
Recession-hit Britons abandon foreign holidays in favour of ‘staycations’ Guardian (13/8/09)
Bad weather puts paid to the Great British Staycation Independent on Sunday (22/8/09)
The following are useful sources of evidence:
Visits to the UK up 4 per cent Office for National Statistics News Release (13/8/09)
1.2 Million More Holidays Taken In England As Brits Take Breaks Closer To Home enjoyEngland (7/8/09)
11.9 million Brits to take U.K break this Bank Holiday enjoyEngland (26/8/09)
- What are the determinants of demand for staycations? How have these impacted on the demand for staycations in the UK in summer 2009?
- How are the (a) price; (b) income and (c) cross-price elasticities of demand for staycations relevant in determining the demand for staycations?
- Why is imperfect information an important problem in making a decision about where to take a holiday and how do risk attitudes affect the decision?
- Why has unemployment risen more than the UK national average in many seaside towns?