
Imagine you run a small restaurant and suddenly a food blogger with 10 million followers posts about you. Your dining room is packed overnight — but you didn't hire more cooks, you just pulled your best people off catering and prep. Revenue shifted between rooms; total output barely moved. That's the core mechanism this paper documents, except the restaurant is a municipality, the blogger is HBO, and the "rooms" are entire economic sectors. The committed claim: a single fictional-TV appearance (Game of Thrones, Season 7, July 2017) caused an abrupt, massive surge in tourism at San Juan de Gaztelugatxe in the Basque Country — but net local employment did not increase because jobs gained in hospitality were offset almost one-for-one by losses in other sectors. The authors call this "sectoral reallocation" rather than growth. They also find housing-market overheating (more transactions, higher prices) and a roughly twofold increase in international arrivals, with no evidence of local population displacement. Methodologically, this is a Synthetic Control study — the workhorse of modern causal inference in regional economics. The authors build a counterfactual version of the treated municipality from a weighted combination of similar, untreated municipalities, then measure how the real site diverged post-shock. The key advantage here is the shock's exogeneity: nobody chose Gaztelugatxe because it was already booming. HBO's location scouts picked it for its geology, not its GDP. That clean identification is what makes the paper publishable. The novel administrative dataset — covering employment, housing transactions, prices, and population — gives them unusually granular municipal-level observation. The ladder context matters. Screen-tourism studies exist — Barcelona after Vicky Cristina Barcelona, Dubrovnik after GoT, New Zealand after Lord of the Rings — but most rely on visitor counts and self-reported surveys, not administrative employment and housing data with causal identification. The authors explicitly note their tourism-activity effects "far exceed" those documented in prior literature, but the economic punchline is the opposite of what the tourism-boosters expect: more tourists ≠ more total jobs. The integrity picture is solid for an observational causal study. Synthetic Control is well-understood, placebo tests are standard practice, and the exogeneity of the shock is about as clean as you get outside a randomized experiment. The main vulnerability is generalizability: Gaztelugatxe is a tiny municipality in a specific institutional context (Basque Country, Spain). Whether the zero-net-employment result holds for larger or differently structured local economies is an open question the paper acknowledges but cannot test. The housing-market finding is the sleeper result. Tourism-driven housing pressure without corresponding employment gains is a politically explosive combination — it describes the mechanism behind anti-tourism backlash in Barcelona, Lisbon, and Venice. If the pattern replicates, it gives policymakers a concrete causal basis for housing-market interventions in tourism-shocked areas. The obvious next experiment is replication across multiple spotlight-shock sites with different labor-market structures — Dubrovnik (GoT), Skellig Michael in Ireland (Star Wars), or Albuquerque (Breaking Bad). The authors almost certainly know this. My read: they're saving the multi-site comparative study for a follow-up, possibly constrained by cross-country administrative data access. The single-site design is the minimum viable publishable unit.
The analogy
Imagine a small restaurant goes viral overnight from a food blogger. The dining room is packed — but you didn't hire more staff, you just pulled people off catering and prep. Revenue shifted between rooms; total output barely moved. That's sectoral reallocation dressed as growth.
Viral restaurant
Blogger spotlight fills the dining room by pulling staff from catering — total labor unchanged, just redistributed
Spotlight tourism
GoT appearance fills hospitality by pulling workers from other sectors — net employment unchanged, just reallocated
Honest assessment
A clean, well-identified single-case study that confirms what urban economists have long suspected: tourism surges reallocate labor rather than creating it, while overheating housing markets. The result is persuasive for this site but needs multi-site replication to become a policy-grade finding. Most useful right now for researchers building the evidence base against naive tourism boosterism, and for policymakers in screen-tourism-affected areas looking for causal evidence to justify housing interventions.
What Is CLAIMS?
70
/ 100
BedrockNews applies a structured analytical framework called CLAIMS to academic papers and technical research releases.
How CLAIMS worksClaim
3/5
Ladder
4/5
Architecture
3/5
Integrity
4/5
Milestone
3/5
Successor
4/5
What is this paper actually committing to?
Real but incremental. First causal identification of tourism's sectoral reallocation mechanism using admin data. Not a new theory — a new empirical confirmation of what urban economists suspected but couldn't prove cleanly.
The field fight
Does tourism growth generate broad-based local prosperity or merely reshuffle existing economic activity?
Side A
Tourism is an export industry that creates net new jobs and income in destination economies
Held by: UNWTO, national tourism boards, booster literature
Side B
Tourism surges reallocate labor across sectors without net gains, while inflating housing costs — a form of Dutch Disease
Held by: Critical tourism scholars, urban economists, anti-overtourism movements
Milestone tracker
Unlocks: Policy-relevant stylized fact: tourism surges reallocate rather than create employment, generalizable across institutional contexts
Current
1
site
this paper
Next
5-10
sites
multi-site replication
~2-4 years, gated by cross-country admin data access
Verified Facts (7)
In July 2017, San Juan de Gaztelugatxe appeared as 'Dragonstone' in Game of Thrones
Employment gains in tourism-related sectors were largely offset by losses elsewhere, resulting in no net effect on total employment
Housing-market overheating documented with increases in both property transactions and prices
No evidence that the shock displaced the local population
Roughly twofold increase in arrivals from abroad
Tourism activity effects far exceeded those documented in the literature
Authors use Synthetic Control approach with novel administrative dataset
Practical Implications (4)
Tourism promotion agencies should track net employment across all sectors, not just hospitality headcounts, when claiming economic benefits from spotlight events
Local governments facing sudden tourism surges should prepare housing-market interventions proactively — the price and transaction effects appear quickly
The zero-net-employment finding gives anti-overtourism movements a causal evidence base rather than just anecdote
Film commissions and location scouts could use this framework to set realistic economic expectations with local stakeholders before granting filming permits
More latest news
Read all