Is there a ‘magic money tree’? Is it desirable for central banks to create money to finance government deficits?
The standard thinking of conservative governments around the world is that creating money to finance deficits will be inflationary. Rather, governments should attempt to reduce deficits. This will reduce the problem of government expenditure crowding out private expenditure and reduce the burden placed on future generations of having to finance higher government debt.
If deficits rise because of government response to an emergency, such as supporting people and businesses during the Covid-19 pandemic, then, as soon as the problem begins to wane, governments should attempt to reduce the higher deficits by raising taxes or cutting government expenditure. This was the approach of many governments, including the Coalition and Conservative governments in the UK from 2010, as econommies began to recover from the 2007/8 financial crisis.
‘Modern Monetary Theory‘ challenges these arguments. Advocates of the theory support the use of higher deficits financed by monetary expansion if the money is spent on things that increase potential output as well as actual output. Examples include spending on R&D, education, infrastructure, health and housing.
Modern monetary theorists still accept that excess demand will lead to inflation. Governments should therefore avoid excessive deficits and central banks should avoid creating excessive amounts of money. But, they argue that inflation caused by excess demand has not been a problem for many years in most countries. Instead, we have a problem of too little investment and too little spending generally. There is plenty of scope, they maintain, for expanding demand. This, if carefully directed, can lead to productivity growth and an expansion of aggregate supply to match the rise in aggregate demand.
Government deficits, they argue, are not intrinsically bad. Government debt is someone else’s assets, whether in the form of government bonds, savings certificates, Treasury bills or other instruments. Provided the debt can be serviced at low interest rates, there is no problem for the government and the spending it generates can be managed to allow economies to function at near full capacity.
The following videos and articles look at modern monetary theory and assess its relevance. Not surprisingly, they differ in their support of the theory!
- Modern monetary theory: the rise of economists who say huge government debt is not a problem
The Conversation, John Whittaker (7/7/20)
- Modern Monetary Theory: How MMT is challenging the economic establishment
ABC News, Gareth Hutchens (20/7/20)
- What is Modern Monetary Theory and is it THE answer?
Sydney Morning Herald, Jessica Irvine (2/7/20)
- MMT: what is modern monetary theory and will it work?
MoneyWeek, Stuart Watkins (14/7/20)
- MMT: the magic money tree bears fruit
MoneyWeek, Stuart Watkins (17/7/20)
- Modern Monetary Theory is no Magic Money Tree
Adam Smith Institute, Matt Kilcoyne (20/5/20)
- “Modern Monetary Theory” Goes Mainstream
Forbes, Nathan Lewis (10/7/20)
- How Boris Johnson’s Conservatives have become Magic Money Tree huggers
The Scotsman, Bill Jamieson (16/7/20)
- Ignore the impacts of debt-fuelled stimulus at your peril
Livewire, David Rosenbloom (14/7/20)
- Modern Monetary Theory, explained
Vox.com, Dylan Matthews (16/4/19)
- Compare traditional Keynesian economics and modern monetary theory.
- Using the equation of exchange, MV = PY, what would a modern monetary theorist say about the effect of an expansion of M on the other variables?
- What is the role of fiscal policy in modern monetary theory?
- What evidence might suggest that money supply has been unduly restricted?
- When, according to modern monetary theory, is a rising government deficit (a) not a problem; (b) a problem?
- Is there any truth in the saying, ‘There’s no such thing as a magic money tree’?
- Provide a critique of modern monetary theory.
Since the financial crisis of 2008–9, the UK has experienced the lowest growth in productivity for the past 250 years. This is the conclusion of a recent paper published in the National Institute Economics Review. Titled, Is the UK Productivity Slowdown Unprecedented, the authors, Nicholas Crafts of the University of Sussex and Terence C Mills of Loughborough University, argue that ‘the current productivity slowdown has resulted in productivity being 19.7 per cent below the pre-2008 trend path in 2018. This is nearly double the previous worst productivity shortfall ten years after the start of a downturn.’
According to ONS figures, productivity (output per hour worked) peaked in 2007 Q4. It did not regain this level until 2011 Q1 and by 2019 Q3 was still only 2.4% above the 2007 Q4 level. This represents an average annual growth rate over the period of just 0.28%. By contrast, the average annual growth rate of productivity for the 35 years prior to 2007 was 2.30%.
The chart illustrates this and shows the productivity gap, which is the amount by which output per hour is below trend output per hour from 1971 to 2007. By 2019 Q3 this gap was 27.5%. (Click here for a PowerPoint of the chart.) Clearly, this lack of growth in productivity over the past 12 years has severe implications for living standards. Labour productivity is a key determinant of potential GDP, which, in turn, is the major limiter of actual GDP.
Crafts and Mills explore the reasons for this dramatic slowdown in productivity. They identify three primary reasons.
The first is a slowdown in the impact of developments in ICT on productivity. The office and production revolutions that developments in computing and its uses had brought about have now become universal. New developments in ICT are now largely in terms of greater speed of computing and greater sophistication of software. Perhaps with an acceleration in the development of artificial intelligence and robotics, productivity growth may well increase in the relatively near future (see third article below).
The second cause is the prolonged impact of the banking crisis, with banks more cautious about lending and firms more cautious about borrowing for investment. What is more, the decline in investment directly impacts on potential output, and layoffs or restructuring can leave people with redundant skills. There is a hysteresis effect.
The third cause identified by Crafts and Mills is Brexit. Brexit and the uncertainty surrounding it has resulted in a decline in investment and ‘a diversion of top-management time towards Brexit planning and a relative shrinking of highly-productive exporters compared with less productive domestically orientated firms’.
- How suitable is output (GDP) per hour as a measure of labour productivity?
- Compare this measure of productivity with other measures.
- According to Crafts and Mills, what is the size of the impact of each of their three explanations of the productivity slowdown?
- Would you expect the growth in productivity to return to pre-2007 levels over the coming years? Explain.
- Explain the underlying model for obtaining trend productivity growth rates used by Crafts and Mills.
- Explain and comment on each of the six figures in the Crafts and Mills paper.
- What policies should the government adopt to increase productivity growth?
A lack of productivity growth has been a major problem for the UK economy over the past decade (click here for a PowerPoint of the chart). Is it possible that the new decade may see a pick-up in the growth in output per hour worked?
One possible solution to low productivity growth is to reduce working hours and even to move to a four-day week, but not to reduce total pay. If people work fewer hours, they may well be more productive in the hours they do work. In fact, not only may output per hour increase, but so too may output per worker, despite fewer hours being worked. What is more, the quality of output may increase with people being less tired and more motivated.
Several companies have experimented with a four-day week, including Microsoft in Japan, which employees 2300 workers. It found that, despite a 20% reduction in hours worked, output per hour worked increased by 40%, with total output thereby increasing. Workers were generally happier and more motivated and asked for fewer days off.
And it is not just a question of output: fewer hours can result in lower costs. The effect on costs will depend on the nature of new work patterns, including whether everyone has the same extra day off.
But a four-day week is only one way of cutting working hours for full-time employees. Another is to reduce the length of the working day. The argument is that people may work more efficiently if the standard working day is cut from eight to, say, five hours. As the first Thrive Global article article (linked below) states:
Just because you’re at your desk for eight hours doesn’t mean you’re being productive. Even the best employees probably only accomplish two to three hours of actual work. The five-hour day is about managing human energy more efficiently by working in bursts over a shorter period.
If people have more leisure time, this could provide a boost to the leisure and other industries. According to a Henley Business School study:
An extra day off could have a knock-on effect for the wider society. We found 54% of employees said they would spend their day shopping, meaning a potential boost for the high street, 43% would go to the cinema or theatre and 39% would eat out at restaurants.
What is more, many people would be likely to use the extra time productively, undertaking training, volunteering or other socially useful activities. Also family life is likely to improve, with people spending less time at work and commuting and having more time for their partners, children, other relatives and friends. In addition, people’s physical and mental health is likely to improve as they achieve a better work-life balance.
So, should firms be encouraged to reduce hours for full-time workers with no loss of pay? Many firms may need no encouragement at all if they can see from the example of others that it is in their interests. But many firms may find it difficult, especially if their suppliers and/or customers are sticking with ‘normal’ working hours and want to do business during those hours. But, over time, as more firms move in this direction, so it will become increasingly in the interests of others to follow suit.
In the meantime, should the government introduce incentives (such as tax breaks) or regulations to limit the working week? Indeed, it was part of the Labour manifesto for the December 2019 election that the country should, over time, move to a four-day week. Although this was a long-term goal, it would probably have involved the use of some incentives to encourage employers to move in that direction or the gradual introduction of limits on the number of hours or days per week that people could work in a particular job. It is unlikely that the new Conservative government will introduce any specific measures, but would probably not want to discourage firms from reducing working hours, especially if it is accompanied by increased output per worker.
But despite the gains, there are some problems with reduced working hours. Many small businesses, such as shops, restaurants and firms offering technical support, may not have the flexibility to offer reduced hours, or may find it hard to increase productivity when there is a specific amount of work that needs doing, such as serving customers.
Another problem concerns businesses where the output of individuals is not easy to measure because they are part of a team. Reducing hours or the working week may not make such people work harder if they can ‘get way with it’. Not everyone is likely to be motivated by fewer hours to work harder.
Then there is the problem if reduced hours don’t work in boosting productivity. It may then be very difficult to reintroduce longer hours.
But, despite these problems, there are many firms where substantial gains in productivity could be made by restructuring work in a way that reduces hours worked. We may see more and more examples as the decade progresses.
- Economics of a four-day working week: research shows it can save businesses mone
The Conversation, Miriam Marra (11/11/19)
- Less hours for work, more time on Earth: Why a four-day working week is good for you
Independent, Steve Taylor (16/12/19)
- Microsoft Japan Launched A Four-Day Workweek To Much Success: Is This The Key To Attracting Talent In The Tight U.S. Job Market?
Forbes, Jack Kelly (5/11/19)
- Will The Five-Hour Work Day Catch On In America?
Forbes, Jack Kelly (28/10/19)
- My Company Implemented a 5-Hour Workday — and the Results Have Been Astounding
Thrive Global, Stephan Aarstol (3/10/19)
- Why Does a Four-Day Work Week Achieve Better Results?
Thrive Global, Stephanie Lin (16/12/19)
- Four-Day Working Week Improves Staff’s Mental Health By 87%, Company Finds
Unilad, Niamh Shackleton (10/12/19)
- Hull business launches four-day working week as well as your birthday off and ‘Beer Fridays’
Hull Live, Phil Winter (5/12/19)
- A four-day work week? Sounds nice, but here’s the real deal
Sydney Morning Herald, Tony Featherstone (21/11/19)
- The four-day debate: Fantasy or feasible?
The Hindu, Business Live, Kamal Karanth (20/11/19)
- Finland’s ministry of transport floats tech-enabled four-day week
Computer Weekly, Gerard O’Dwyer (20/11/19)
- Should You Consider a 4 Day Work Week?
Small Business Trends, Rob Starr (5/12/19)
- Distinguish between different ways of measuring labour productivity.
- Give some examples (from the linked references) of employers which have tried introducing a four-day week or reduced hours for full-time workers. What has been the outcome in each case?
- In what ways may reducing working hours reduce a firm’s total costs?
- What are the advantages and disadvantages of the government imposing (at some point in the future) a maximum working week or a four-day week?
- What types of firm might struggle in introducing a four-day week or a substantially reduced number of hours for full-time employees?
- What external benefits and costs might arise from a shorter working week?
A general election has been called in the UK for 12 December. Central to the debates between the parties will be their policy on Brexit.
They range from the Liberal Democrats’, Plaid Cymru’s and Sinn Féin’s policy of cancelling Brexit and remaining in the EU, to the Scottish Nationalists’ and Greens’ policy of halting Brexit while a People’s Vote (another referendum) is held, with the parties campaigning to stay in the EU, to the Conservative Party’s policy of supporting the Withdrawal Agreement and Political Declaration negotiated between the Boris Johnson government and the EU, to the DUP which supports Brexit but not a version which creates a border between Great Britain and Northern Ireland, to the Brexit Party and UKIP which support leaving the EU with no deal (what they call a ‘clean break’) and then negotiating individual trade deals on a country-by-country basis.
The Labour Party also supports a People’s Vote, but only after renegotiating the Withdrawal Agreement and Political Declaration, so that if Brexit took place, the UK would have a close relationship with the single market and remain in a customs union. Also, various laws and regulations on environmental protection and workers’ rights would be retained. The referendum would take place within six months of the election and would be a choice between this new deal and remain.
But what are the economic costs and benefits of these various alternatives? Prior to the June 2016 referendum, the Treasury costed various scenarios. After 15 years, a deal would make UK GDP between 3.4% and 7.8% lower than if it remained in the EU, depending on the nature of the deal. No deal would make GDP between 5.4% and 9.5% lower.
Then in November 2018, the Treasury published analysis of the original deal negotiated by Theresa May in July 2018 (the ‘Chequers deal’). It estimated that GDP would be up to 3.9% lower after 15 years than it would have been if the UK had remained in the EU. In the case of a no-deal Brexit, GDP would be up to 9.3% lower after 15 years.
When asked for Treasury forecasts of the effects of Boris Johnson’s deal, the Chancellor, Sajid Javid, said that the Treasury had not been asked to provide forecasts as the deal was “self-evidently in our economic interest“.
Other forecasters, however, have analysed the effects of the Johnson deal. The National Institute for Economic and Social Research (NIESR), the UK’s longest established independent economic research institute, has estimated the costs of various scenarios, including the Johnson deal, the May deal, a no-deal scenario and also a scenario of continuing uncertainty with no agreement over Brexit. The NIESR estimates that, under the Johnson deal, with a successful free-trade agreement with the EU, in 10 years’ time UK GDP will be 3.5% lower than it would be by remaining in the EU. This represents a cost of £70 billion. The costs would arise from less trade with the EU, lower inward investment, slower growth in productivity and labour shortages from lower migration. These would be offset somewhat by savings on budget contributions to the EU.
Under Theresa May’s deal UK GDP would be 3.0% lower (and thus slightly less costly than Boris Johnson’s deal). Continuing in the current situation with chronic uncertainty about whether the UK would leave or remain would leave the UK 2% worse off after 10 years. In other words, uncertainty would be less damaging than leaving. The costs from the various scenarios would be in addition to the costs that have already occurred – the NIESR estimates that GDP is already 2.5% smaller than it would have been as a result of the 2016 Brexit vote.
Another report also costs the various scenarios. In ‘The economic impact of Boris Johnson’s Brexit proposals’, Professors Anand Menon and Jonathan Portes and a team at The UK in a Changing Europe estimate the effects of a decline in trade, migration and productivity from the various scenarios – again, 10 years after new trading arrangements are in place. According to their analysis, UK GDP would be 4.9%, 6.4% and 8.1% lower with the May deal, the Johnson deal and no deal respectively than it would have been from remaining in the EU.
But how much reliance should we put on such forecasts? How realistic are their assumptions? What other factors could they have taken into account? Look at the two reports and at the articles discussing them and then consider the questions below which are concerned with the nature of economic forecasting.
- UK’s new Brexit deal worse than continued uncertainty – NIESR
Reuters, David Milliken (30/10/19)
- Brexit deal means ‘£70bn hit to UK by 2029′
BBC News, Faisal Islam (30/10/19)
- Boris Johnson’s Brexit deal worse for economy than Theresa May’s, new analysis shows
Politics Home, Matt Honeycombe-Foster (30/10/19)
- Boris Johnson’s Brexit deal ‘would cost UK economy £70bn’
The Guardian, Richard Partington (30/10/19)
- UK economy suffers ‘slow puncture’ as general election is called
ITV News, Joel Hills (30/10/19)
- Boris Johnson’s Brexit deal ‘would deliver £70bn hit to economy by 2029’
Sky News, Ed Conway (30/10/19)
- Boris Johnson’s Brexit deal won’t cost Britain £70bn by 2029
The Spectator, Ross Clark (30/10/19)
- Boris Johnson’s Brexit deal would make people worse off than Theresa May’s
The Guardian, Anand Menon and Jonathan Portes (13/10/19)
- How Boris Johnson’s hard Brexit would hit the UK economy
Financial Times, Chris Giles (13/10/19)
- Boris Johnson’s Brexit deal is worse for the UK economy than Theresa May’s, research suggests
CNBC, Elliot Smith (19/10/19)
- What are the arguments in favour of the assumptions and analysis of the two recent reports considered in this blog?
- What are the arguments against the assumptions and analysis of the two reports?
- How useful are forecasts like these, given the inevitable uncertainty surrounding (a) the outcome of negotiations post Brexit and (b) the strength of the global economy?
- If it could be demonstrated beyond doubt to everyone that each of the Brexit scenarios meant that UK GDP would be lower than if it remained in the EU, would this prove that the UK should remain in the EU? Explain.
- If economic forecasts turn out to be inaccurate, does this mean that economists should abandon forecasting?
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.