,
17 tweets,
7 min read
Read on Twitter
My top three thoughts on the "aha, but agglomeration benefits don't actually exist! Gotcha!" piece in The Guardian.
1/ People are better off in Blackpool than in Manchester (local authorities).
2/ If you control for all the causes of differences in productivity, they go away.
1/ People are better off in Blackpool than in Manchester (local authorities).
2/ If you control for all the causes of differences in productivity, they go away.
3/ Agglomeration effects barely work in the UK, so I'm not sure why controlling was necessary.
4/ I'll get and read the actual paper as soon as possible. The Guardian have written up my work before, they never spoke to me and I disagreed with most of what had been written.
4/ I'll get and read the actual paper as soon as possible. The Guardian have written up my work before, they never spoke to me and I disagreed with most of what had been written.
Very grateful that the paper is open access from day one. shura.shu.ac.uk/24893/
My first impression on reading this paper is of how wonderfully clear it is. It is really rare to read something like this. I suspect it's partly because it's not been peer-reviewed and published in a journal. Everything is set out clearly, the examples are excellent.
The size of the commuter effect adjustments to GVA/head for the UK's big cities and London surprises me. This feels like a bigger correction than I've used in the past -- I wonder if it's partly an artefact of using LEP geographies and not a more functional geography?
The result is as expected though -- we know that agglomeration benefits for the UK's main regional cities just don't work. The UK's towns and countryside are no less prosperous than the UK's cities once London is excluded. This is unlike all other big developed countries.
Yet again, the quality of the presentation of the findings in this work is exceptional. I may not agree with what the authors claim their results show, but I can see exactly what they've done, I could easily try and repeat the work, and I can follow their logic very easily.
And now the end. A table of the results, so that people repeating the calculations can see if they're doing it right. I may try and write an alternative "Implications for public policy" section, but it would be consistent with the data and calculations in the piece. Excellent.
I am glad to have read to the paper -- I found the write-up in The Guardian really poor; this is much better. As I say, I'm not convinced by all of the policy implications that stem from this work, and I will just check that I can repeat the calculations. But this is ace work. 👍
I now see why the commuter effect is much bigger than I expected. It's because on page 17 we move from looking at LEPs to looking at the UK's big regional cities as defined by local authority. And so of course Manchester has huge inward commuter flows.
One tweet summary,
The paper: "agglomeration benefits don't exist in the UK so maybe government shouldn't invest as if they do".
My reply: "agglomeration benefits don't exist in the UK (except London) because government hasn't invested as if they could, it should now start".
The paper: "agglomeration benefits don't exist in the UK so maybe government shouldn't invest as if they do".
My reply: "agglomeration benefits don't exist in the UK (except London) because government hasn't invested as if they could, it should now start".
There's an additional element to the paper which is "low productivity areas aren't low productivity because people are lazy" --- which of course I agree with completely. I rarely hear someone argue otherwise to me, and I never hear them argue otherwise twice.
I also reject the argument (the existence/popularity of which in academia I find somewhat astonishing) that low productivity parts of the UK have low productivity because the "good/intelligent" people left and only the "bad/stupid" people remain. tomforth.co.uk/onintelligence/
That @Paul_Swinney at @CentreforCities has also read the paper. I agree with him. The best examples are those within South-East England, where agglomeration benefits definitely do exist within the UK. centreforcities.org/blog/has-the-m…
A very positive outcome of this paper, which has surprised me a bit, is that lots of Brits seem to have learned that the books they read about the US economics (where agglomeration benefits exist and the bigger a city is the more productive it is) are not applicable to the UK.
Except for London there is good evidence that agglomeration benefits do not work in the UK. Almost every agglomeration booster I know agrees. The @CentreforCities have a slide on it in almost every presentation (here's a recent example). As do I.
And here are my slides. Four in the most recent presentation!
