In recent years, shoppers in Türkiye do not need the evening news to tell them prices have been rising, because they see it at the till. But the till only tells you about the history of inflation. When they turn on the television, the news might tell them what to expect next, and which channel they choose will matter.
Orkun Saka (City St George's, University of London) and his co-authors matched a monthly household survey to the transcripts of prime-time news on 13 Turkish channels. He tells Tim Phillips about how the way a channel talks more about inflation, and how often it mentions the word, affects what viewers expect will happen next. Do polarised TV channels lead to a "polarisation of reality"?
The research behind this episode:
Demiralp, Selva, Orkun Saka, and Michael Weber. 2026. "Polarized Media and Inflation Expectations." CEPR Discussion Paper 21747. Paris and London: Centre for Economic Policy Research.
To cite this episode:
Phillips, Tim, and Orkun Saka. 2026. “Inflation and the polarisation of reality.” VoxTalks Economics (podcast).
About the guest
Orkun Saka is a Reader in Economics at City St George's, University of London. He is also affiliated with the London School of Economics and the CESifo Network. His research spans financial intermediation, international finance and political economy.
Research cited in this episode
Turkey's 2022 inflation peak. Annual consumer price inflation in Turkey reached 85.5% in October 2022, according to the Turkish Statistical Institute, its highest level in 24 years. It followed a run of interest rate cuts while inflation was rising. Orkun uses the episode to explain why expectations in Türkiye became so detached from the official figures. See Al-Monitor (2022).
Heterodox monetary policy. For several years the Turkish government argued that lower interest rates would bring inflation down, the reverse of the textbook view. In the experiment, respondents were asked whether they agree that "lower interest rates are needed to curb high inflation"; neutral-channel viewers whose inflation expectations rose became less likely to agree.
Freedom House press rating. In 2014, Freedom House downgraded Türkiye's press from "partly free" to "not free" in its annual Freedom of the Press report. The paper uses this, with evidence on media ownership and regulation, as background to its classification of 13 channels as pro-government, neutral or pro-opposition.
Koç University Household Inflation Expectations Survey (KUHIES). Koç University. 2024. Koç University Household Inflation Expectations Survey. A nationally representative monthly survey of about 2,500 Turkish adults, launched in January 2024. It asks respondents what inflation was over the past 12 months, what it will be over the next 12, which channel they watch for prime-time news and how they voted in 2023. In the paper's sample, mean expected inflation was 101%.
News selection as a signal. Nimark, Kristoffer P., and Stefan Pitschner. 2019. "News Media and Delegated Information Choice." Journal of Economic Theory 181: 160-196. Newsrooms decide what is important enough to report, so the choice to cover a topic tells the audience something before they hear a word of the story. Orkun's finding that the amount of inflation coverage matters more than its tone fits this idea.
The 2024 local elections. Turkey held local elections on 31 March 2024. In the run-up, election news crowded economic news out of prime time, so inflation coverage fell for reasons unrelated to viewers' beliefs. The authors use this as a natural experiment; as coverage fell, expectations fell with it.
Bayesian updating. A way of describing how people revise beliefs; they start from a prior, see new information, and arrive at a posterior that weighs the two. People with weak priors move further when new information arrives. In the experiment, the test is whether the usual close link between a respondent's prior and posterior inflation expectation weakens after reading a news story.
Experimenter demand effects. de Quidt, Jonathan, Johannes Haushofer, and Christopher Roth. 2018. "Measuring and Bounding Experimenter Demand." American Economic Review 108 (11): 3266-3302. Survey participants may give the answer they think the researcher wants. To reduce this, the experiment asked about posterior expectations using a differently worded question from the first.
The polarisation of reality. Alesina, Alberto, Armando Miano, and Stefanie Stantcheva. 2020. "The Polarization of Reality." AEA Papers and Proceedings 110: 324-328. People with different political views hold systematically different beliefs about the same facts. Orkun argues that inflation is a particularly worrying place for this to happen, because monetary policy depends on a shared view of where prices are going.
More VoxTalks Economics episodes
How should we measure expected inflation? Ricardo Reis on which measures of inflation expectations are reliable, and how well they anticipated the surge in European inflation from 2021.
Social media polarization. Yves Zenou on when social media pulls opinions apart and when it brings people together, the open question Orkun raises at the end of this episode.
Related reading on VoxEU
When the news you watch shapes the inflation you expect, a VoxEU column summarising the panel and experimental evidence from this research.
Household inflation expectations: Taking stock of the recent research insights for monetary policy, a VoxEU column by Francesco D'Acunto, Evangelos Charalambakis, Dimitris Georgarakos, Geoff Kenny, Justus Meyer and Michael Weber on how households form and act on inflation beliefs.
Monetary policy in the news: The FOMC's media coverage and inflation expectations, a VoxEU column showing that hawkish media coverage of the Federal Reserve lowers households' inflation expectations, especially when inflation is high.