Economic possibilities for our grandchildren, video games version

Famously, John Maynard Keynes predicted in 1930 that in a few generations time people would only work 15 hours a week: productivity would have risen so much that higher living standards would be possible with less work.

He thought that people would use higher productivity (and the resultant higher pay per hour) to work much less, and consume much more leisure. But that didn't quite happen: labour hours did fall, but much much slower than he expected, despite productivity growing about as much as he thought it would. People wanted more consumer goods, as well as more services, than he thought was likely.

He (and his followers, the Skidelskys) thought it was an appreciation for the higher pleasures—like contemplation and philosophy—that would eventually take up leisure time.

But it seems that it is artificial reality, in the form of ever higher quality video games, that is the first use of leisure tempting enough to really stop men working, at least according to the work of Erik Hurst & collaborators. Here's the abstract of their new paper:

Younger men, ages 21 to 30, exhibited a larger decline in work hours over the last fifteen years than older men or women. Since 2004, time-use data show that younger men distinctly shifted their leisure to video gaming and other recreational computer activities.

We show that total leisure demand is especially sensitive to innovations in leisure luxuries, that is, activities that display a disproportionate response to changes in total leisure time. We estimate that gaming/recreational computer use is distinctly a leisure luxury for younger men.

Moreover, we calculate that innovations to gaming/recreational computing since 2004 explain on the order of half the increase in leisure for younger men, and predict a decline in market hours of 1.5 to 3.0 percent, which is 38 and 79 percent of the differential decline relative to older men.

See also two Marginal Revolution posts on the phenomenon, from Tyler Cowen and Alex Tabarrok. I seem to remember a whole load of folks attempting to ridicule them for believing in this at the time, but the evidence does seem to be getting stronger and stronger.

Sorry Adonis, but there's no tuition fee cartel

On Friday, Lord Adonis had an op-ed published in The Guardian in which he conducts a volte face concerning tuition fees. Specifically, although he was responsible for increasing tuition fees from £1,000 per year to £3,000 per year, he now believes that they should be scrapped entirely on the (actually spurious) grounds that graduating students now leave university with about £50,000 of debt.

Notwithstanding the lack of a cogent argument for scrapping tuition fees, Adonis makes the highly-charged allegation that universities are running a cartel because a large number of them set fees at or close to the £9,000 maximum that is permitted. This allegation shows a shocking lack of understanding of the meaning of the term “cartel” and the jurisprudence underlying previous instances in which cartels were found to be operating.

In short, a cartel (as per Article 101 of the Treaty on the Functioning of the European Union) is defined as a group of firms that restrict or distort competition in a market by, for example (but not limited to), directly or indirectly fixing prices, limiting or sharing production / output, and so on. As such, Adonis is claiming that the universities are fixing prices when they all set their fees close to £9,000 and claims that he has asked the Competition and Markets Authority (CMA) to investigate this matter.

However, Adonis ignores the established and standard guidance for determining whether or not a cartel is actually in operation – simply taking the fact that a number of universities are pricing at a focal point is not evidence in and of itself. Instead, as per the guidance set out by the then Court of First Instance in Airtours/First Choice, three conditions are necessary for a cartel finding.

  1. The market must be transparent – i.e. the area (such as prices, volumes etc.) over which a cartel agreement is made must be able to be monitored easily and (relatively) costlessly for the members of the cartel. In the case of university pricing, this condition probably is satisfied as a university’s fees are published on their website and therefore can be monitored by other universities. However, it is the only one of the three conditions that is satisfied.
  2. More importantly, the members of the cartel must be able to ”punish” any cartel member that does not adhere to the cartel agreement. This is one area where Adonis’ claim falls down – there is no punishment mechanism available for universities to punish those that deviate from a cartel agreement. To see this, suppose that there was an agreement between universities to maintain fees at £9,000. Now suppose that one university that had made this agreement instead decided to cut its fees to £4,000 – this “deviating” firm might have a strong incentive to do so since that could enable it to get more students applying to it, enabling the university to have a wider range of students from which it could select the best candidates.

    In this scenario, how would the non-deviating universities be able to punish the deviator? They couldn’t decrease their own fees because 1) those fees are set some time in advance so cutting them as a rapid response is not really feasible; and 2) that would simply decrease those universities’ own revenues anyway, without the likely prospect of recouping that loss in future, such that doing so would be cutting off their nose to spite their face. Moreover, since the impact of the punishment would arise a year later (i.e. when the next set of students were applying to university), the punishment itself would not provide a strong disincentive to prevent the deviator from cutting fees in the first place.
  3. Any “external” competition (i.e. options other than going to the universities in the supposed cartel) must be weak. Again, Adonis’ argument fails on this criterion too - although the vast majority of UK universities are charging fees at the upper limit, the fact is that there are multiple outside options that provide a competitive constraint: 1) private universities are increasing in size and coverage and are often a viable alternative for students; 2) prospective students can choose not to go to university, but instead attend a technical college or other institution; and 3) future students can choose to study in foreign universities and are doing so in ever greater numbers. In other words, there are external options that constrain the ability of UK universities to cartelise fees.

Therefore, universities in the UK do not seem to satisfy two of the three conditions required for a cartel to be present. Hence, it is highly likely that Lord Adonis’ claim that UK universities are operating a cartel is baseless and will be given short shrift by the CMA.

We're never going to get things right if The Observer believes nonsense like this

We're quite obviously closer to the classical liberalism of the Observer's past than the progressive liberalism of their present so disagreements over policy are going to happen. But we shouldn't find that we've got disagreements on fact for unless we can all both discover and agree upon those we're never going to get anything right.

Sadly, we do so disagree:

The first task is to end the absurdity of bogus self-employment. This employment category is plainly being abused by some companies whose workers continue to be told how to behave like regular employees for all practical purposes – their work can be directed down to exact details, including time, hours, place and uniform. The only real difference is that the employer is able to avoid paying employers’ national insurance and pension contributions, and offering protections such as maternity and sick pay. It’s a scam. Massive benefits accrue to the employers and very few to the workers.

We agree entirely that all of those things are lovely for the workers to have. If they desire them of course. But we would insist that it's not the employers paying for them, all such benefits are incident upon the wages of the workers. No, really, even Richard Murphy agrees with us here.

There is an amount which employers are willing to pay for the labour being used. Whether this is sliced into wages and tax, or wages and tax and a pensions contribution and national insurance doesn't matter, the amount is the amount. Increasing the amount that must be paid in not-wages thus decreases the amount paid in wages, not the total amount being paid.

The absence of these "benefits" does not benefit the employer therefore, it raises the workers' wages. And imposing these costs will, over time at least, lower said workers' wages even as their compensation stays static.

We might wonder why there has been this explosion of jobs paying wages only, no benefits. And then we might look at the minimum wage, which insists that wages must be at a certain nominal level. To keep total compensation at the amount that the employers are willing to pay the non-wages part shrinks as the law insists that there's a floor to wages themselves.

What else did anyone think was going to happen? As cash wages have risen the non-wage part of compensation has fallen to compensate. And note what will happen next if we insist that those non-wage benefits must exist. That minimum wage floor of cash wages will start to bite and we'll get that unemployment as we raise the total compensation above what the job is worth.

All of this stemming from the simple fact that employers' NI, like any other non-wage cost of employment, is incident upon the workers, not the employers. And we really do think that a national newspaper, however progressively liberal it is, should be able to grasp this simple fact.

The problem is that what Trevor Nunn believes about inequality isn't true

Trevor Nunn is touting this new play he's involved with which is - wait for it - on the subject of the gross inequality of today. The problem being here that what he believes about inequality just isn't true

And yet, even in our enlightened social democratic western world, we remain utterly unequal – probably more so now than at any previous time.

This simply isn't so. Not in any sense that matters of course. It's well known that two of the three richest people on the planet, Bill Gates and Warren Buffett, are partial (Lord knows why) to a Big Mac on occasion. There are almost none of us in the rich world who do not have equal access to those. Nothing by Maccy D's might not be healthy but it's financially out of reach of very few of us. And it's extremely doubtful that this has even been true before, that the entire society has equal access to the food desired by the richest. The plebes didn't get access to those lark's tongues after all.

That is, of course, being tendentious, as is pointing out that we've all entirely equal access to Facebook and WhatsApp on the same terms.

Yet when we do these calculations properly, as the TUC once did, we do indeed find that the only form of inequality that matters, that of consumption, is low, very low. Using quintiles of households the TUC found that the 5 th (ie, the top) earned 12 times the 1st. After we subtracted taxes, added benefits, levered in the value of public services like the NHS and education, the consumption inequality came down to 4 to 1.

It is extremely difficult to think of any previous version of human society which was as equal as this.

Not that the inequalities which remain are also of rather less import. Not too long ago poverty meant no shoes, now it means off brand sneakers. Inequality meant empty bellies, unto the point of death - yes starvation existed in England up into the 19th century. Now such inequality, or if you prefer, poverty,. appears to mean a greater propensity to obesity.

I have to presume that the motive for acquisition is competition: that driving force, that god in Margaret Thatcher’s universe, the market. Competition to defeat all your rivals, competition to be able to declare you have more wealth than anybody except Bill Gates, competition to get more than Bill Gates.

Which is also grossly wrong. Competition is what limits that wealth that can be amassed. Monopolists do tend to be rich, the competition from Steve Jobs, Linus Torvalds, Larry Ellison and all the rest is precisely what has limited the fortune of Bill Gates.

No doubt the play will be a success given how many share these delusions. But it is a delusion, we are more equal today than almost all human societies since the invention of agriculture.

The terrors of the patriarchy in tech

A worry expressed in The Guardian. AI more generally, the use of big data to train it more specifically, risks coding into the systems the attitudes of the current society:

But this can create problems when the world is not exactly as it ought to be. For instance, researchers have experimented with one of these word-embedding models, Word2vec, a popular and freely available model trained on three million words from Google News. They found that it produces highly gendered analogies. For instance, when asked “Man is to woman as computer programmer is to ?”, the model will answer “homemaker”. Or for “father is to mother as doctor is to ?”, the answer is “nurse”. Of course the model reflects a certain reality: it is true that there are more male computer programmers, and nurses are more often women. But this bias, reflecting social discrimination, will now be reproduced and reinforced when we engage with computers using natural language that relies on Word2vec. It is not hard to imagine how this model could also be racially biased, or biased against other groups.

Yes indeed, that will happen. If you train something (anything, this applies to guide or gun dogs as much as it does to a translation program) to operate within what exists then it will operate by the rules which currently exist.

The important part here being of course that part about the world not exactly as it ought to be. This starts to smack rather of New Soviet Man, that luscious planned economy would start working right around the time that we've changed humans so that they work within that luscious planned economy. Which isn't, quite, how it turned out, was it? 

But perhaps it should be different this time? To which there is an answer, the answer being the same even as the questions differ. We use markets and competition to work this out. Some section of society would prefer to see all such structured genderism not incorporated into those AIs. Other (very few perhaps) would like to see more of it, most possibly just wanting it to reflect the real world not the dreamed one. All of which is absolutely fine. There is no shortage of capital out there, there are no legal of cultural constraints upon anyone building an AI absolutely any way they wish to.

We'll find out which people really prefer when they use the various available alternatives and, well, use them. That's what we've done with every other invention and innovation in history after all and look how much better off we are than those societies which didn't - New Soviets for example.  

At which point:

Products that are more responsive to the needs of women would be a great start. 

Well, get on with it then. For surely you're not insisting that the men should do it for you, are you?

So just why are companies sitting on vast lakes of cash?

It's not an unusual complaint these days, companies sit on vast lakes, reservoirs' worth, of cash and don't seem to do much with it. Something must have changed so what is it? At least part of the answer, accounting for just under a third of it, is that the world of business has changed. Well, obviously, but we've identified one of the ways that it has

I explore the role of the just-in-time (JIT) inventory system in the increase in cash holdings by U.S. manufacturing firms. I develop a model to illustrate the mechanism through which JIT affects cash and quantify its impact. In the model, both cash and inventory can serve as working capital. As firms switch from the traditional system to JIT, they shift resources from inventory to cash to facilitate transactions with suppliers. On average, this switchover accounts for a 4.1-percentage-point increase in the cash-to-assets ratio, which is approximately 28% of the change observed in the data.

Back decades a company might sit on months of inventory, today perhaps hours. That needs less working capital to finance the inventory of course, but also more ready cash to pay suppliers. What happens to the net investment position depends upon the balance of those two effects.

Our own guess at it, and it is a guess, would be that the reduced working capital demands outweigh the greater liquidity demands. Thus rather neatly explaining something else people seem to worry about, the decline in corporate use of capital itself.

There is also a larger point to make here, we must always be very sure that we are distinguishing cyclical changes from structural. It is possible, for example, that people might use corporate demands for cash or working capital as a measure of the state of the economy. OK, fine, why not do so? But when we do so we've got to make sure that as the underlying technology, and thus the structural demands, change we do not confuse the two, the structural and the cyclical.

Our favourite example of this error was a measure of business software investment. That this is static was used to argue that there was no great technological revolution going on and therefore we faced secular stagnation. The problem being that business increasingly rents its software, not buys it. Office and Office 365 both do the same thing, but Office is an investment, 365 a current expense. Business was indeed therefore "investing" more in software, by the amount of the size of the software as a service market, but this was completely missed by the use of the investment statistic.

Or, as we like to say, it's no use looking at an economic statistic unless you understand, in detail, what it is actually measuring and why

The Fiction of the ‘Great Capital Rebuild’

The central element of the Bank of England’s narrative on the UK banking system is the ‘Great Capital Rebuild’. To paraphrase Governor Carney’s comments when the 2015 stress tests were released: the post-Global Financial Crisis (GFC) period and the long march to higher capital are over. The message – which he has repeated since – is that UK banks are now more or less fully capitalised.

Unfortunately, the ‘Great Capital Rebuild’ is a fiction.

Let’s look at the evidence. Exhibit A is the following chart (Chart B.2) from the BoE’s November 2016 Financial Stability Report.

Major UK Banks’ Leverage Ratios

Sources: PRA regulatory returns, published accounts and Bank calculations.  (a)   Prior to 2012, data are based on the simple leverage ratio defined as the ratio of shareholders’ claims to total assets based on banks’ published accounts (note a discontinuity due to introduction of IFRS accounting standards in 2005, which tends to reduce leverage ratios thereafter). The peer group used in Chart B.1 also applies here. (b)   Weighted by total exposures. (c)  The Basel III leverage ratio corresponds to aggregate peer group Tier 1 capital over aggregate leverage ratio exposure. Up to 2013, Tier 1 capital includes grandfathered capital instruments and the exposure measure is based on the Basel 2010 definition. From 2014 H1, Tier 1 capital excludes grandfathered capital instruments and the exposure measure is based on the Basel 2014 definition.

Sources: PRA regulatory returns, published accounts and Bank calculations. 

(a)   Prior to 2012, data are based on the simple leverage ratio defined as the ratio of shareholders’ claims to total assets based on banks’ published accounts (note a discontinuity due to introduction of IFRS accounting standards in 2005, which tends to reduce leverage ratios thereafter). The peer group used in Chart B.1 also applies here.

(b)   Weighted by total exposures.

(c)  The Basel III leverage ratio corresponds to aggregate peer group Tier 1 capital over aggregate leverage ratio exposure. Up to 2013, Tier 1 capital includes grandfathered capital instruments and the exposure measure is based on the Basel 2010 definition. From 2014 H1, Tier 1 capital excludes grandfathered capital instruments and the exposure measure is based on the Basel 2014 definition.

This chart shows some of the BoE’s own estimates of UK banks’ leverage ratios spanning 2001 to 2016: the leverage ratio is the ratio of some measure of capital to the total amount at risk. This chart indicates that UK banks’ leverage ratios are a little higher than a decade ago – maybe 25% on this measure, but certainly no multiple – and a decade ago the banks were on the eve of an almighty crash.

Now comparing leverage ratios before the GFC and after is a tricky business because of definitional changes made by Basel III. Yet the Bank itself publishes figures for two leverage ratios known as Simple Leverage Ratios (SLRs): the ratio of shareholder equity to total assets. One refers to the book value of shareholder equity and the other to the market value of shareholder equity. These series give average SLRs across the banking system and span the period from before the GFC until recently. [2] To the extent that we can rely on these to give us a before and after comparison, the average book value SLR was just under 4.1 percent in 2006 and 6.2 percent in the first half of 2016, representing an increase of 51 percent. [3]

The corresponding market value SLR was 8.0 percent going into 2006 and 5.28 percent in November 2015, representing a decrease of 34 percent. [4] By this latter measure, UK banks are more highly leveraged now than they were going into the crisis.

I would suggest that it would be prudent to pay attention to these market value figures: the market values being less than book values is a signal that the market perceives problems with the book values. 

Then consider that the big four banks’ total Common Equity Tier 1 (CET1) capital was about £205 billion by the end of 2016q3. This figure is barely £90 billion higher than the £116 billion Tier 1 capital that they had going into 2007, although one must acknowledge that this £90 billion difference does not allow for the considerable improvement in quality between Basel II Tier I capital and Basel III CET1.

The 2016q3 £205 billion CET1 number is a book value figure, however, and the corresponding market value of its CET1 capital was about £149 billion.

We should also assess these numbers against the sizes of the banks’ balance sheets, and it is traditional to use Total Assets as such a measure. Given that their Total Assets were just under £5 trillion at the same date, their average CET1 leverage ratio (or ratio of CET1 capital to total assets) was 4.0 percent if we go by book values and just under 3 percent if we go by market values.

By the first measure, UK banks are leveraged by a factor of 1 divided by 4 percent or 25: they have £25 in assets for every £1 in capital; and by the second measure, they are leveraged by a factor of over 33. These are high levels of leverage that leave the banks vulnerable to shocks – and high levels of leverage aka inadequate capital were a key factor contributing to the severity of the GFC. 

Putting all this together, the evidence for a ‘Great Capital Rebuild’ is not there – especially if one pays attention to the market value numbers.

As a further confirmation, Chart B.3 in the Bank’s November 2016 Financial Stability Report states that “Most capital rebuilding to date has reflected falls in risk-weighted assets” – a delightful piece of duckspeak – and then gives a breakdown of this ‘rebuild’ in terms of its constituent components. The rebuild it is referring to is not quite what it might seem, however: it refers to the rebuild in the banks’ average ratio of CET1 capital to risk-weighted assets (RWA) relative to 2009. Now the CET1 ratio was 6.92 percent in 2009 and had risen to 12.61 percent by end-2015. That increase breaks down into 0.45 percentage points in new equity raised, 1.02 percentage points in retained earnings and 4.22 percentage points in reductions in risk-weighted assets. Therefore, only 1.47 percentage points of that increase in the capital ratio represents actual increases in capital; the rest, the 4.22 percentage points decrease in risk-weighted assets merely reflects the decrease in the denominator. I would suggest that the chart should have stated “Most of the increase in the ratio of capital to RWAs to date has reflected falls in risk-weighted assets” but that doesn’t quite convey the same message. The increase in the capital ratio from 6.92 percent to 12.61 percent might seem impressive at first sight – an increase of 82 percent – but the actual capital rebuild was only from 6.92 percent to 8.39 percent, an increase of about 21 percent.

A big increase in a regulatory capital ratio is one thing, but a big increase in actual capital is quite another.

Let’s face it: the ‘Great Capital Rebuild’ is not there in the data.

[1] Kevin thanks Sir John Vickers for helpful inputs to this posting.

[2] These figures will overstate the leverage ratio and understate true levels of leverage because they use the larger Shareholder Equity measure rather a narrow core capital measure such as Core Tier 1 or CET1, but they give some sense of the trends over time.

[3] These figures are to be found on p. 57 of the Bank’s November 2016 Financial Stability Report.

[4] These figures come from the BoE Excel workbook ‘ccbdec15.xlsx, spreadsheet ‘9. Bank equity measures’ under the C column, ‘Market-based leverage ratio (%)’. This workbook was accessed on March 9th 20616 but appears to have been removed from the BoE website since the time I accessed and downloaded it.


This is how it works, first you redefine poverty then you find lots of it

Over the decades the definition of poverty has changed, from meaning someone not having very much to someone having less than others. That move from measuring poverty to relative poverty turns it from a measure of actual poverty into one of inequality. The reason this was done is obvious, Britain abolished absolute poverty in the 1930s and where would the left be without something to whine about? 

We are fortunate to be witnessing an attempt to again redefine:

More than a million people in the UK experienced destitution in 2015, including 312,000 children, according to a groundbreaking study by Heriott Watt University academics for the Joseph Rowntree Foundation published last year. It defined destitution in two ways: experience of at least two of six poverty measures over the previous month, including eating fewer than two meals a day for two or more days; or a weekly income after housing costs of £70 for a single adult or £140 for a couple with children. This was an income level below which people “cannot meet their core material needs for basic physiological functioning from their own resources”.

A definition of destitution which defines it as only two meals a day twice a month is rather co-opting a word with a very strong meaning to cover something really quite different. 

We should also note, as we so often do, that this definition is before the things which people other than the state do to alleviate such distress.

What we are seeing here is an attempt to float another redefinition, like that to relative poverty measures, in order to show that there is some vast problem which needs more tax to solve.