Refugees and credible commitments

A couple of days ago The Telegraph published an article by Suzanne Evans (the Ukip deputy chairman) concerning whether or not the UK should agree to take the 3,000 unaccompanied children that are currently in the Calais jungle. The main argument is embodied in the paragraph from the article:

Take 3,000 from the jungle and 3,000 more will arrive within days. Then another 3,000 will arrive days after that. How many will die on the journey? How many will be terrified, starved, subjected to life-threatening diseases and horribly abused on the way to France? How many of the older kids will be exposed to hard drugs while they’re in the camp, which appears to be run by louts at best and hardened criminals at worst? The fact is, the more we take in, the more children will be abandoned to the cruel sea and the even crueler people traffickers. Is this really what we want?

In other words, Evans is saying that if the UK was to accept the current 3,000, then that would only encourage more families to make the journey, with subsequent risks to those families.

A family’s decision whether or not to attempt the journey to the UK comes down to a balance between “pull” factors (e.g. the chance of getting into the UK) and “push” factors (e.g. the danger involved in not making the journey and remaining in their home countries). Evans assumes that the UK accepting 3,000 children would materially alter that decision. If we take that assumption as given, then Evans also assumes that the UK cannot credibly commit not to take any more children after the current 3,000.

What does “credibly committing” to something mean? A credible commitment occurs when an entity makes a believable and enforceable promise to do (or not do) something. This is often discussed within the field of monetary policy where a central bank cares about having lower inflation and lower unemployment. If the central bank simply announces that inflation will be lower, then once people belief that inflation will be lower, the central bank has an incentive to renege on the announcement so as to obtain lower unemployment at the expense of higher inflation. Hence, the original announcement would not be believed in the first place, preventing the central bank from obtaining lower inflation.

However, if the central bank could come up with a way for it to “credibly commit” to keeping inflation low, then the public would believe its announcement, such that inflation could be decreased. How to achieve such a credible commitment? In the case of a central bank, an inflation target is one potential way.

But what relevance does this have to migrants making a perilous journey, you might ask?

As mentioned, one assumption implicit in Evans’ argument is that the UK cannot credibly commit to saying “after these 3,000 children we won’t accept any more”, thereby meaning that other families would make a dangerous journey to try to get to Calais in the hope that they too would eventually be accepted into the UK. However, if the UK was able to credibly commit to taking such action (the form that such a credible commitment could take would be for various politicians to decide), then migrants choosing between making the dangerous journey or not would have less of an incentive to make said journey.

Moreover, a credible commitment by the UK might not necessarily affect families’ decisions at all. If the “push” factors are so strong, a slight change one way or another in the “pull” factors via a UK credible commitment is unlikely to substantially decrease the number of families making the journey.

As such, it seems that whether or not the UK accepts the current 3,000 children in the Calais jungle is highly unlikely to have a substantial impact in encouraging additional families to try to make a similar journey.

Advertisements

(Not) Eating for two: Fasting and educational attainment

Recent news regarding the possibility of moving the dates on which core GCSE exams are held forward so as to avoid Ramadan has highlighted the impact that fasting and nutrition can have on exam results and educational attainment.

The impact of fasting / poor nutrition during exam periods and/or while growing up has been subject to numerous studies before – see, for example, Anderson et al. and  Glewwe & Miguel.

However, what has received less attention (at least until now) is the impact of a mother’s fasting during pregnancy on a child’s educational attainment. That is why a paper by Douglas Almond, Bhashkar Mazumder, and Reyn van Ewijk in the latest issue of The Economic Journal is rather interesting – it looks at precisely this issue.

Specifically, the paper looks at the impact of Ramadan falling during pregnancy (i.e. the impact of fasting during the gestation period) on the educational attainment of Pakistani and Bangladeshi children (i.e. those children most likely to be of Muslim decent and, hence, whose mothers are most likely to have fasted during Ramadan) at age seven compared to that of other groups of children and finds that there is a small, but significant, decrease in educational attainment of the Pakistani and Bangladeshi children. The authors use this result to suggest that “brief prenatal investments may be more cost effective than traditional educational intervention in improving academic performance.

On the whole, the paper is a good one, following a clearly set out method, and the results provide useful policy indications. However, that is not to say that the paper does not have some flaws.

First, the paper uses the educational attainment of children of Caribbean decent as the “control group” against which the educational attainment of Pakistani and Bangladeshi children is tested. The authors justify this on the grounds that Caribbean families are unlikely to fast during Ramadan, such that the control group is unaffected by Ramadan. Although this might seem reasonable from a statistical perspective (although I’d argue that it still introduces unacceptable biases into the analysis), from a policy perspective it is less desirable.

Specifically, the relevant control group to determine whether a policy would be worthwhile is “the average student” – the authors do not provide any evidence to indicate that Caribbean students represent the average attainment in the UK. Indeed, the paper actually suggests that Caribbean students’ educational attainment tends to be below the UK average. In other words, the effect that the paper finds is likely to be understated relative to the average student, such that the paper’s conclusions could be much stronger if the average educational attainment was used. (Although the paper conducts a “robustness check” using White British students as an alternative control group, this still fails to get at the effect compared to the average student.)

Second, the paper assumes a “standard” gestation period of xx days, but does not investigate the extent to which changing the length of this gestation period affects the results. It could be that a slight change in the length of the gestation period assumed by the authors would affect the effect they find, which is important given that the effect they find is relatively small (albeit statistically significant). Hence, the rigour of the paper could have been improved by including this sensitivity check.

Third, the authors fail to make sufficient inference from the results presented in the paper. In particular, the paper’s results indicate that the impact of fasting on educational attainment differs according to the stage during the gestation period at which the fasting occurred – the impact of gestational fasting on educational attainment is largest when it happens during the third and fourth month of gestation and is almost negligible when it occurs after the seventh month. In other words, the paper could have highlighted the importance of early nutritional interventions for policy during gestation, but failed to do so.

Nonetheless, despite these flaws, the paper is an interesting one, with some important policy implications.

Statistics, state benefits, and reproduction: some sort of unholy trinity

Over the past week, there has been some discussion regarding a book written by Adam Perkins (a Lecturer in the Neurobiology of Personality at King’s College London) regarding the impact of state benefits on personal outcomes (the title of the book – “The Welfare Trait: how state benefits affect personality” – probably gives that away).

Of particular focus has been one table that Perkins claims demonstrates that being on benefits is associated with a higher number of children in that household – i.e. that being on state benefits encourages reproduction (according to Perkins). The table, reproduced from here below, appears to indicate that households in which no-one is employed have higher numbers of children than do households with one or two workers.

[Children%2520per%2520household%255B4%255D.png]

Perkins’ conclusions were criticised by Jonathan Portes (a Senior Fellow at the NIESR) for a number of reasons, to which Perkins provided a response – see here.

However, this response is entirely unsatisfactory as it fails to acknowledge the actual problems with using the table above to make meaningful inferences. First, it does nothing to demonstrate that the results in the table are statistically significant – that is to say that the results in the are not just due to “random chance” (i.e. an artefact of the data used) and are, in fact, a “true” result. Without testing the results for statistical significance, there is no way to determine whether or not there is actually a meaningful difference between the number of children in working households as compared to workless household. Indeed, Perkins does not appear to even acknowledge this as an issue of his reliance on this table.

Second, Perkins’ numbers deliberately exclude households in which there are no children. This is an egregious decision that Perkins has tried to justify using baseless arguments. The issue is that Perkins could have deliberately introduced a bias into the numbers in the table that directly affects the inferences drawn from the results in the table. To see this, suppose that a higher proportion of workless households have no children than the proportion of working households that have no children.

Indeed, suppose that 500,000 working households do not have children, but 1,000,000 workless households do not have children. The table below, shows the impact of this hypothetical example – including households that do not have children actually reverses the direction. Once childless households are included in this hypothetical example, workless households actually have fewer children that do working households. Hence, it is essential that the number of childless households are included when trying to see if state benefits affect reproduction.

Hypothetical childless

In fact, when one uses the actual total number of households available from the ONS (rather than the hypothetical example above), we actually find a mixed set of results. The number of children per workless household is indeed lower than the number of children in working and mixed households, but the number of children per mixed household is higher than that in working households.

Actual childless

Hence, Perkins’ conclusions regarding the impact of benefits on reproduction are incorrect. The actual results suggest that there does not appear to be a systematic relationship between the employment status of a household and the number of children in that household. This goes to show how excluding certain groups from a dataset can introduce bias in any results obtained from analysing that dataset.

Corbyn and the railways: the costs of having a life-size trainset

A major part of Corbyn’s campaign to become leader of the Labour Party was the promise to re-nationalise the operation of the UK railways. Indeed, that promise became Corbyn’s very first official policy, with part of the supposed rationale being that “the public have paid £10 billions in subsidies and the operators have posted aggregate profits of £1 billion” since the railways were privatised.

Although there have been a few articles in the mainstream media that purport to examine the feasibility of re-nationalisation in terms of the costs of doing so (see, for example, here and here), these articles have, at best, been cursory and only scratched the surface.

In particular, these articles fail to examine 1) the initial costs of regaining control of the railways from private operators; and 2) the annual costs of running the railways once they have been re-nationalised. These are discussed in turn below.

The first issue of the government regaining control of the railways could be solved costlessly (in terms of government money) if a Corbyn government were willing to wait until the franchises running the railways came to a natural end. However, this approach does not seem likely for two reasons. First, the current schedule of rail franchises indicates that all will run out prior to 2020 (the first year of an hypothetical Corbyn government), such that the current Conservative government likely would re-new the franchises at lengths of around 7-10 years (as suggested by the Brown Review) means that the earliest date at which a franchise will expire under an hypothetical Corbyn government would be 2022. Second, given that the policy of re-nationalisation was the first of Corbyn’s official policies, it seems that Corbyn would be unwilling to wait particularly long to enact this policy.

Hence, Corbyn’s re-nationalisation most likely would require buying out all franchise owners. Although there do not appear to be any publicly available figures regarding how much it would cost to buy out all rail franchises, we can construct a back-of-the-envelope estimate. Using the fact that Stagecoach and Virgin paid £3.3bn for one franchise (implying that the franchisees consider a franchise to be worth at least that much), and assuming that this value is directly related to the number of passenger kilometres (i.e. the value of a passenger kilometre is the same across all franchises, then using the information provided in the Office for Rail Regulation Report here (excluding London Overground),  the total value of all rail franchises in the UK is about £38bn. This amounts to a one-off cost of roughly 3% of GDP in the year in which the Corbyn government would re-gain control of the railways.

Although that doesn’t sound too bad, there is also the second item involved in running the railways – namely, the ongoing costs (or profits) from operating them. Indeed, currently the franchises together have a profit margin of about 3% – in other words the railway operators make roughly £250 million per year as profit.

While it might be tempting to conclude that re-nationalising the railways would therefore actually make it profitable on an annual basis to do so, this would imply that it would take almost 150 years for the initial £38bn cost to be recouped by the government. That probably does not make much business sense.

Moreover, that £250 million per year figure assumes that the current prices, price increases, and levels of investment apply in future. However, isn’t the entire rationale for Corbyn wanting to re-nationalise the railways so that prices are kept lower and investment higher than they would be if the railways were not re-nationalised? Although it is difficult to put an exact figure on how much it would cost per year to achieve Corbyn’s aims of lower prices and/or higher investment levels, it is plausible that they would pretty much wipe out any current profit. In other words, the initial outlay of £38bn is unlikely to ever be recouped.

Now, whether you think that this outlay of public funds that are unlikely to ever be recouped (and that could be used for other things such as new hospitals, new schools etc. if they were not used to re-nationalise the railways) is worth having possibly (but not guaranteed) lower prices and higher investment levels, then that’s your choice. But I know where I’d rather have £38bn spent, and it’s not on taking back control of the railways.