What is the epidemiological impact of a #falsenegative#COVID test? An important question in a high vaxx/low NPI context, but one that cannot be studied in a experiment for obvious reasons. Enter the UK, a reliable supplier of #naturalexperiments. 🧵⬇️
➡️bit.ly/3DhqQv5
On Oct 15, @UKHSA suspended an #Immensa lab, because of community reports of neg PCR tests following a pos lateral flow. There was loads of excellent reporting e.g. by @rowenamason@tomjs@JamieGrierson. NHS TT estimates that 43,000 individuals may have been given a .. 2/N
false negative result most concentrated in South West of England. Even across all of England, a notable increase in both absolute # and relative % of PCR tests matched to a positive LFD tests producing a negative result from early Sept to early Oct 3/N
this is also visible in overall testing comparing the 13 allegedly most affected districts with the rest of England, you see similar evolution of % of positive tests & # of cases, up until 2 Sept from then on the rates go down, before shooting up after the lab was suspended 4/N
Geographically, map plots the % of positive tests A. four weeks prior to the lab returning false neg; B. the 5 weeks when it was producing false negs; C. & four weeks after. D plots highlights districts in SW standing out with massive ⬆️ in % tests after suspension of lab.
control approach focusing on the 13 districts that were said to have been most affected. This allows for a cleaner quantification of the impacts by contrasting case figures relative to a more credible/cleaner counterfactual 6/N
Doing so, I estimate that every missed case has lead to between 0.6 to 1.6 additional infections. Combined, this would imply that 43k missed + cases implied between 25 - 68k additional cases. 7/N
The analysis is also done district-by-district, suggesting some heterogeneity. But really more data is needed to refine the analysis as it does suggest that the % of tests processed by the Wolverhampton lab varied over time. 8/N
The lumpiness of missing cases on specific dates suggests that errors may have been due to faulty equipment, environmental conditions or consistent mishandling in specific days/shifts. The @Independent@samueljlovett had a piece with a whistleblower independent.co.uk/news/health/co… 9/N
Now where to take it from here? It's imperative that a full independent investigation is carried out of what went wrong. Maintaining trust in the testing system & the guidelines is key, but so is accountability and transparency. The pandemic has been a financial boon to many 10/N
many firms that were awarded huge contracts without tender. The UK testing system is much better than what exists in e.g. Germany, but it has cost taxpayers around GBP 37 bn. So its evident there needs to be more public scrutiny & its crucial academics get the right data 11/N.
Glad to see that my evidence submission to UK Parliament Committee on Use of AI in Government has been quoted in todays report. There is much said and some case study examples in the piece. I want to use this 🧵to sketch out policy implications of expanded use of AI in govt...
Implication 1: With AI + digital payment infrastructure, tax filing/compliance cost cost can vastly decrease allowing for a broader fiscalisation that could be implemented with a reduction in the VAT registration threshold that is very high in international comparison in UK.
Implication 2: Citizens perceive poor front-line services and bloated middle management that can be automated. Significant need to reallocate workforce a) out of offices to front line services & b) to become integral part of the innovation ecosystem.
@EconMaett @infornomics I think also tagging @edenhofer_jacob here. I think the big challenge continues to be that, of course, foreign interference is entirely possible on social media in the absence of (digital) ID. We simply dont know if it is humans or bots or networks of bots operating accounts.
The problem is that the issue of ID is a transnational one. States are traditionally providing identity layers; issue currencies; etc. The right of free speech is vital, but it does not come with the right to an audience. And in social media space, you can easily "fake" an audience. This is part of the hybrid attack on institutions. But the US is caught up in wanting to protect the economic interests of its tech companies and has used this as a wedge issue within Europe finding varying coalitions. In the end there is no way around it that some form of capital controls will come back as these are now technologically feasible.
the "attack" on some business models of large tech companies that basically enable structural CIT tax evasion (think: transfer pricing) is an EU case and so its not surprising that tech companies covered and protected by US policy (as happened with Trump v1 tax cuts) fight back and it may well be that their best strategy is to break up the EU.
This is a longer thread to connect some dots. It is so obvious to some to understand what "went wrong" in 2000s/2010s but as I keep saying deeds matter more than words. I will try to relate some academic work that speaks to this. This is also about understanding...
The above is a status report on the G20 Data Gaps Initiative (DGI) tackling data deficiencies to improve policy-making. The Phase 3 started in, surprise surprise, 2022. Focus areas are 1️⃣ Climate Change 2️⃣ HH Distributional Info 3️⃣ Fintech/Fin Inclusion 4️⃣ Data Access/Sharing
Anybody who is somewhat on top of things would recognize these as immediate priorities to allow mechanisms like CBAM, carbon credit trading, compensation etc. to work and to embed this into a national accounting framework. Informational capacity needs to be explicitly be built...
How do spatially skewed economic shocks deepen gender employment gaps? New research by @sarthak_joshi who is on the market this year reveals that rising Chinese imports reshaped labor demand in India, disproportionately restricting women’s access to urban non-farm jobs...
due to gendered commuting frictions.
Main finding: Improving transport infrastructure for women could have:
✅ Mitigated female labor force declines by 30% (2001–2011)
✅ Boosted total output by 0.4%
A stark case for policy intervention to empower women and grow economies.
This is obviously relevant to all the great research that has document the distributional implications of trade shocks more broadly highlighting the role of fiscal policy accommodating and shaping these shocks to cushion or shape their impact. In the context of the UK & the US...
Interesting to read Iversen and Rosenbluth's book, especially in light of more recent research on gender equality. I'll highlight some of my favourites below.
Just to echo Jacob's point: Read @PikaGoldin!!!
Goldin, Claudia. 2006. ‘The Quiet Revolution That Transformed Women’s Employment, Education, and Family’. American Economic Review 96(2): 1–21.
--. 2014. ‘A Grand Gender Convergence: Its Last Chapter’. American Economic Review 104(4): 1091–1119. doi:10.1257/aer.104.4.1091.
--. 2023. ‘Why Women Won’. doi:10.3386/w31762.
Here are some fascinating papers by economists:
@kuhnmo et al
Want to stop paying for Github copilot? Dont wait any longer. Here a simple step-by-step instruction on how you can integrate any LLM trained on code into your workflows.
Step 2: If you dont have already, download and install VScode (my preferred coding environment right now) and then
Step 3: install the continue extension