Not all eggs are created equal

Afternoon all,

Good news Britain – we’ve had a little ‘the economy did better than thought’ pick me up from the ONS this morning (the UK managed growth of 0.3 per cent in Q1 2023 up from the previous estimate of 0.1 per cent). Combine that with the earlier upgrade to the size of the immediate pandemic bounce back in 2021 and we now look ‘normal’ by European standards.

Normal means roughly the same post-Covid recoveries as France and Italy, but ahead of Germany (being more economically dependent on Russia/China is… suboptimal right now). Obviously, everyone this side of the pond is lagging North America (the last two years have been easier if you produce oil/gas, rather than being used to importing it from Russia).

Once you’ve stopped celebrating you can get on with the serious business of this week’s reads and we’ve got a great COTW showing why “eggs is eggs” is… increasingly nonsense.

Have a great weekend.

Torsten

Gleeful graduates. Apparently ignorance is bliss (bet you didn’t know that line came from a poem about Eton), but does education really make us gloomy? We know education improves your earning and your health. On the latter front read Anne Case and Angus Deaton’s latest note on life expectancy in the US: there’s now an 8.5 year gap in adult life expectancy between those with and without a degree  – a huge rise from 2.6 years in the early 90s. And a new paper finds that more education generally means higher life satisfaction, but with one important catch: the relationship inverts for unemployed people, amongst whom having more education decreases life satisfaction. The authors explanation? Unmet aspirations (for income or status) are bad for your life satisfaction. Just ask Laurence Fox.

Accurate AI? There’s so much AI nonsense out there, but a new note (summary) is worth your time because it elegantly makes a simple point: to understand the strengths/weaknesses of large language models (currently centre stage in AI developments) it’s important to remember what they are (algorithms trained on internet content to predict what the next word in a sequence is) and aren’t (replicating human thought processes/learning). So LLMs are much better at providing answer if those answers (or indeed the questions) are common online. What does this mean? Well ChatGPT is okay (right half the time) at multiplying by 9/5 and adding 32, but terrible at near identical task of multiplying by 7/5 and adding 31. Why? Because only the former is common on the interweb (to convert temperatures from Celsius to Fahrenheit). The authors conclude that we should not treat these large language models as humans. Who was treating them as humans?!

Ruinous recruitment. I did not see this paper coming, but it turns out there’s nothing economists won’t analyse… including Islamic State’s personnel records. The authors examine nearly 4,000 recruits who travelled to Syria/Iraq in the middle of the last decade, where they know their country of origin and qualification level. The conclusion? Higher unemployment in countries near to Syria saw more foreign fighters recruited (1 percentage point lower unemployment in those countries would have meant 1,200 fewer recruits joining Daesh). For countries further away the relationship between unemployment and recruitment inverted, reflecting higher travel costs acting as a barrier to reaching the war zone.

Impeding infrastructure. Lots of HS2 this week – chat that is, not actual train line. Obviously. The revived will we/won’t we finish the thing, amidst not unrelated cost overruns, have prompted a row about why it always costs more to build in Britain than elsewhere. There’s some mixed evidence about how true/significant that is, but ignore that nuance because it definitely costs too much so we should all be focusing on why the costs can get so high. Sam Dumitrui (who has done great work in this area) new blog has a good canter through the potential drivers that others have put forward, from onerous planning, to nasty NIMBYs, costly consultants, and problems with level/volatility of public investment (I’ve tweeted X-ed my contribution on that front). Action on all of these would be desirable. Bargain building is what we need.

Healthy habits. Collapsing NHS + money is tight = at some point surely we’ll start focusing on a healthier population being part of the answer. The cost of living crisis means taxes on unhealthy food are out of fashion (despite the success of George Osborne’s sugary drinks tax), so what about healthy food subsidies? They generally increase healthy foods consumption (subsidising things tends to do that…). But more useful hints on how to do this well comes via a US experiment examining how such subsidies are delivered with two findings. First, offering people a degree of agency in opting into a healthy focus for the subsidy (vs a baked good cake subsidy) actually increases overall fruit and veg consumption (the 20 per cent of us opting for the cakes are outweighed by higher fruit and veg consumption by the virtuous 80 per cent). Second, giving people time to forward plan healthy shopping (i.e. informing them of subsidies in advance) pays off by apparently putting us in a ‘future-focused mindset’. In other news I’m helping run the cake sale at the kids’ school in a few hours…

Chart of the week

Eggs is eggs is increasingly not true. At least price wise. COTW crams in decades worth of economic and social change, examining not only the typical price of eggs but their distribution over time. Obviously typical egg prices are up since 1998 (yes I know the farmers amongst you will say not by enough) due to the whole inflation thing. But what’s interesting isn’t that prices have drifted up but that the *variation* in egg prices is up massively. In Thatcher’s final years in No10 the poshest 10 per cent of eggs typically set you back £1.26, with the cheapest costing only £0.78. Today? There is far more price variety. The poshest eggs (£5.85) have pulled far ahead, while the typical dozen (£3.70) remains closer to cheapest eggs (£2.55). This is part of a wider story of product choice increasing, as we get luxury (and own brand) versions of almost everything. The angsty take on egg inequality is that it exemplifies the rich pulling away from the middle and bottom – they don’t even dip soldiers into the same yolks any more! The other side of the argument? Maybe £5 a dozen super organic eggs are a much more cunning device for soaking the rich than inheritance tax…