Chris Cook

Carmen Reinhart and Ken Rogoff have had a bad day. The two economic historians’ research, which implied that public debt overhangs can hamper economic growth, was perhaps one of the most cited pieces of work in recent years. Their advice that high debt-GDP ratios – particularly above 90 per cent – are harmful to growth, has become a widely used point in discussion. And it’s under attack by a trio at the University of Massachusetts, Amherst – Thomas Herndon, Michael Ash, and Robert Pollin.

As FT Alphaville has noted, the issue is about one of Reinhart and Rogoff’s most heavily cited papers on the importance of debt. This paper has been accused of being the victim of fat-fingered Excel coding, as well as selective use of data and odd weighting of how different episodes are weighted, which seemed – to the authors – to make little sense. [UPDATE, 14:46 GMT - Pollin and Ash have written a piece for the FT on what they think this all means for austerity here.]

Robin Harding posted Reinhart and Rogoff’s original reply here. Overnight, the authors have worked through the numbers – and have put up a pretty robust defence of their work. They do admit the first error – there was an Excel blunder:

…Herndon, Ash and Pollin accurately point out the coding error that omits several countries from the averages in figure 2. Full stop. HAP are on point. The authors show our accidental omission has a fairly marginal effect on the 0-90% buckets in figure 2. However, it leads to a notable change in the average growth rate for the over 90% debt group.

They are, however, resisting the second issue – the selective use of data.

HAP go on to note some other missing debt data points, which they describe as “selective omissions”. This charge, which permeates through their paper, is one we object to in the strongest terms. The “gaps” are explained by the fact there were still gaps in our public data debt set at the time of this paper.

They also defend the weightings:

Our approach has been followed in many other settings where one does not want to overly weight a small number of countries that may have their own peculiarities

And they stand by their conclusions.

So do where does this leave matters on debt and growth? Do Herndon et al. get dramatically different results on the relatively short post war sample they focus on? Not really. They, too, find lower growth associated with periods when debt is over 90 per cent. Put differently, growth at high debt levels is a little more than half of the growth rate at the lowest levels of debt.

To address that point, here are some tables from Prof Reinhart showing the difference between their work in the American Economic Review in 2010 and the debunking, using data from 1945 to 2009. The new research shows some linkage between debt and growth.

Reinhart & Rogoff Herndon, Ash & Pollin
Debt/GDP Mean Median Mean Median
<30 4.1 4.2 4.2 NA
30-60 2.8 3.9 3.1 NA
60-90 2.8 2.9 3.2 NA
>90 -0.1 1.6 2.2 NA

She enclosed another table, with a longer timespan, back to 1800:

Reinhart & Rogoff Herndon, Ash & Pollin
Debt/GDP Mean Median Mean Median
<30 3.7 3.9 NA NA
30-60 3 3.1 NA NA
60-90 3.4 2.8 NA NA
>90 1.7 1.9 NA NA

Prof Reinhart also pointed out that an article of theirs, published in the Journal of Economic Perspectives in 2012, found countries with a debt ratio of below 90 per cent had average growth rates of 3.5 per cent – compared to 2.4 per cent for those above 90 per cent – for 1800 to 2011.

Their full response is below the fold. Read more

For the past two years, the Financial Times has had exclusive access to the National Pupil Database, which provides anonymous exam performance data for every individual secondary school pupil in England.

Using this detailed data, we have compiled an alternative set of statistics to the traditional measures, which are now set to be reassessed. Read more

Chris Cook

Later this morning, Michael Gove, education secretary, will announce several big things. First and foremost, he is dropping his plan to introduce the EBC, his proposed new qualification for 16 year-olds, which has been attacked as fatally flawed since its announcement. Second, he will unveil details of the new curriculum. Both will deservedly absorb lots of column inches.

But Mr Gove will also announce a new pair of measures by which league tables will be constructed. This change might actually be the most important thing he does during his entire reign. League tables set out the incentives that drive schools. They define success and failure.

So what do we know? Schools will, first, be assessed on the share of pupils getting Cs or better in English and maths. A second measure will record whether children in each school do better or worse than children of similar ability – as measured by standardised tests at the age of 11.

This value-added score will gauge performance across English and maths, as well as three more core subjects and their three best ‘other’ subjects. This replaces the current measure – a crude tally of how many children get Cs or better in English, maths and three other subjects.

 Read more

Chris Cook

There is an iron law in English education: as any given argument about any problem with schools progresses, the probability that someone will claim grammar schools are the solution rapidly tends towards 1.

I thought I would set out the data on the grammar counties, where children are sorted at the age of 11 according to an academic test.

To do this, I have defined a new region of England: Selectivia. I have removed the biggest selective counties – Kent, Lincolnshire, Medway and Buckinghamshire – from their geographical regions and shoved them together into one new region*. So what is it like? First, you can see that this region is quite well off, compared to most regions, especially London.

Region IDACI score FSM
East Midlands 0.195 12.0%
East of England 0.168 9.2%
London 0.340 22.4%
North East 0.245 17.4%
North West 0.233 16.2%
Selectivia 0.162 8.8%
South East 0.150 8.3%
South West 0.164 9.4%
West Midlands 0.236 16.4%
Yorkshire and the Humber 0.216 14.6%

 Read more

Chris Cook

It is fairly well established, as various people have pointed out over the past few days, that poor children in the UK are more likely to be overweight than their richer peers. This is often seen as a curious reversal of older norms: poor children used to be lean.

But one aspect of modern poverty is the same as ever. Inner city school leaders sometimes talk about children looking poorer than others. What they are referring to is not weight, but height. Poor kids are usually shorter (especially ex-refugees). Read more

Chris Cook

Today is league table day, when school exam results are published. The most interesting part of the table is the bottom: 195 schools* are below the government’s “floor targets”. These schools are risk of being taken over by a third party to turn them around (if the process is not already underway).

Schools in this category have fewer than 40 per cent of their pupils get Cs or better in English, maths and three other subjects. They must then also have fewer than 70 per cent of the schools’ pupils making “expected progress” in both English and maths.

A few system-level observations:

  • London does really well. Really well. Only 11 of its schools are below target. Only four are in inner London. The outer boroughs are now a bigger educational problem than the inner city.
  • None of the 164 selective schools was below the floor. Grammar schools cruise to the floor target, because they select bright kids. But some might repay a visit by the inspectors: three of them made less-than-expected progress in English.
  • About one third of failing schools are sponsor academies already. The DfE has yanked on the convert-to-academy lever a lot already. But there are another 132 schools left below the floor target not already attached to sponsors.
  • The academy chains are not going to find it easy to take them on. The best academies are all in London. The worst schools are not. There is limited really good improvement capacity in chains outside the capital, where it is needed.
  • The converter academies were not all good schools. Already, there are 14 which are not meeting the standard. The DfE’s terror of sorting out struggling academies is going to become an ever-bigger problem.

And here is the data: first, what type of schools are below the floor target. For the neophytes, voluntary-aided and -controlled schools are the two types of English religious schools. Foundation Schools and CTCs are types of schools with more independence from their local authorities than others (both are precursors to the academies). Studio Schools and UTCs are types of employer-led school (see here for more on them):

School type Above target Below target
Sponsor academies 249 63
Converter academies 666 14
CTCs 3
Community schools 897 72
Free schools 5
Foundation schools 475 33
Studio Schools 1
UTCs 2
Voluntary aided 399 11
Voluntary controlled 64 2

Second, this is where the struggling schools are. I have broken this down by the GCSE-age school population, which brings out some of the variation in regional school quality more clearly.

GCSE pupils in above-target schools GCSE pupils in below-target schools Proportion of pupils affected
East Midlands 41,799 4,490 9.7%
East of England 56,847 4,054 6.7%
London 71,817 1,746 2.4%
North East 25,450 1,782 6.5%
North West 70,592 4,802 6.4%
South East 78,976 5,113 6.1%
South West 50,789 1,736 3.3%
West Midlands 57,328 3,194 5.3%
Yorkshire and the Humber 51,500 4,359 7.8%

 Read more