Joël Cariolle

@joelcariolle

Researcher #econsky #devecon #digital @fondationferdi.bsky.social & @cerdi. Anti-anti-woke. Currently interested in digital and cable-related stuffs. Music lover, DJ. 🇺🇦🇵🇸 Views are personal. Website:

882 Siguiendo612 Seguidores139 Publicaciones

Bluesky

¿Es tu perfil?Reclámalo y joelcariolle.dirty.social es tuyo.Reclamar perfil

Joël Cariolle@joelcariolle· 20d

Post a perfect album from the 90's that isn't nirvana, soundgarden, pearl jam, or alice in chains

Joël Cariolle@joelcariolle· 22d

🔎 𝗜𝗺𝗽𝗼𝗿𝘁𝗮𝗻𝘁𝗹𝘆, 𝗵𝗶𝗴𝗵-𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗲𝗱𝗻𝗲𝘀𝘀 𝗰𝗼𝘂𝗻𝘁𝗿𝗶𝗲𝘀 — with + connected partners, diversified cable ownership structure and #IXPs — 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲 𝗻𝗼 𝘀𝗶𝗴𝗻𝗶𝗳𝗶𝗰𝗮𝗻𝘁 𝗹𝗼𝘀𝘀𝗲𝘀. Policy levers for #digitalresilience are therefore well identified #DigitalVulnerability #DigitalDisasters

Network redundancy as a moderator of GDP losses.
Notes: CDH event-study estimator (de Chaisemartin and d’Haultfoeuille 2024). Countries split at the sample median, groups fixed
at baseline values. Red hollow markers: above-median redundancy. Blue filled markers: below-median redundancy. Specifications
include country fixed effects, year fixed effects, country-specific linear trends, and region×year and income-group×year fixed effects.
Standard errors clustered at the country level; 95% confidence bands. Source: Cariolle (2026).

Joël Cariolle@joelcariolle· 22d

📈 𝗕𝘂𝘁 𝗻𝗲𝗮𝗿𝗯𝘆 𝗻𝗼𝗻-𝗱𝗶𝘀𝗿𝘂𝗽𝘁𝗲𝗱 𝗲𝗰𝗼𝗻𝗼𝗺𝗶𝗲𝘀 𝗴𝗮𝗶𝗻: they expand bandwidth-intensive service exports and attract higher FDI inflows, with TFP and productivity rising in mirror image. Considering those spillovers, the net growth gap between exposed and unexposed countries reaches ~7 p.p.

Cable breaks and competitiveness, direct effects and regional spillovers.
Notes: Figures report event–time coefficients with 90% confidence bands. All specifications include at least country fixed
effects, year, and income group × year fixed effects, a linear country trends, with standard errors clustered at the country
level. Top-row graphs include region×years fixed effects. FDI inflows per capita are transformed using the inverse hyperbolic
sine. Labor productivity levels are measured as real GDP in US dollars measured in 2010 prices and exchange rates, divided
by employment. TFP is a Human capital-adjusted TFP growth metric, calculated as a residual of labor productivity growth
by subtracting the contribution of human capital and capital deepening to labor productivity growth. Year 7 corresponds to a
period of six years after the cable failure occurrence. The regional treatment is either the number of affected economies or the
number of disruptions –since one disruption can affect multiple economies– recorded in the region of the non-affected country.

Joël Cariolle@joelcariolle· 22d

💡 𝗥𝗲𝘀𝘂𝗹𝘁𝘀 📉 𝗔 𝘁𝘆𝗽𝗶𝗰𝗮𝗹 𝗱𝗶𝘀𝗿𝘂𝗽𝘁𝗶𝗼𝗻 𝘁𝗿𝗶𝗴𝗴𝗲𝗿𝘀 𝗮𝗻 𝗶𝗺𝗺𝗲𝗱𝗶𝗮𝘁𝗲 1-2% 𝗱𝗿𝗼𝗽 𝗶𝗻 𝗚𝗗𝗣 𝗽𝗲𝗿 𝗰𝗮𝗽𝗶𝘁𝗮, 𝘄𝗶𝗱𝗲𝗻𝗶𝗻𝗴 𝘁𝗼 𝗮 𝟵% 𝗱𝗶𝘃𝗲𝗿𝗴𝗲𝗻𝗰𝗲 𝗼𝘃𝗲𝗿 4–5 𝘆𝗲𝗮𝗿𝘀. These connectivity shocks translate into synchronous contractions in private consumption, imports, exports, FDI, and productivity.

Event-study coefficients on the 2008–2019 sample (2020 excluded). Top row: GDP per capita growth (%) for cable
disruptions (left) and repair days (right). Bottom row: log GDP per capita for cable disruptions (left) and repair days (right).
Specifications include country and year fixed effects, linear country trends, and nonparametric region and income-group trends,
with standard errors clustered at the country level. 95% confidence bands. Year 7 corresponds to six years after the failure;
the medium-run magnitude is anchored at t + 4–t + 5.

Joël Cariolle@joelcariolle· 22d

Yet the costs of abrupt connectivity losses remain far less understood. 𝗪𝗵𝗲𝗻 𝗮 𝗰𝗮𝗯𝗹𝗲 𝗯𝗿𝗲𝗮𝗸𝘀, 𝗳𝗶𝗿𝗺𝘀 𝗰𝗮𝗻'𝘁 𝘃𝗲𝗿𝗶𝗳𝘆 𝗼𝗿𝗱𝗲𝗿𝘀, 𝗽𝗮𝘆𝗺𝗲𝗻𝘁𝘀 𝘀𝘁𝗮𝗹𝗹, 𝗮𝗻𝗱 𝗱𝗮𝘁𝗮-𝗶𝗻𝘁𝗲𝗻𝘀𝗶𝘃𝗲 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀 𝗴𝗿𝗶𝗻𝗱 𝘁𝗼 𝗮 𝗵𝗮𝗹𝘁. Effects compound through supply chains, banking systems, and investor confidence.

Mapping of countries affected by the SEA-ME-WE and FLAG systems failure, December 2008

Joël Cariolle@joelcariolle· May 14

Joël Cariolle@joelcariolle· Apr 25

Interesting & useful thx! I had a paper rejected but with good feedbacks from the editors of AER, Restat, and good reports from JEEA. But it looks the LLM overstated it (and made a misjudgement), or maybe it should also consider corresponding author's university in the eval 😜:

Joël Cariolle@joelcariolle· Mar 23

Very happy to be part of this session on agriculture at the @oxfordcsae.bsky.social 2025 conference. Please join us today tone the riverside theatre from 11:00 to 13:00 gmt. #csae2025

Agriculture 1 session’s topics and lineup at the CSAE 2025 conference

Joël Cariolle@joelcariolle· Oct 27

Codes bsky en vente sur le bon coin... Non mais sans déconner !🤯