Thread by @asadliaqat

Hi #EconTwitter! I’m a PhD candidate at Harvard @Kennedy_School on the market. Here's my JMP:

No⏰?🇵🇰Politicians know little about what citizens prefer. Better information makes them more responsive, especially towards women.

Lots of⏰? bit.ly/2KyR7vt

Some⏰? See⬇️

BIG Q for development = What affects public good provision?

We have excellent evidence on many determinants: corruption, red tape, uninformed voters + others

I explore a new one: info gaps on the politician's side

Q: Do politicians know enough about citizens to represent them?

📢I first establish that politicians have highly inaccurate beliefs about what citizens prefer: they are only marginally more accurate than a random guesser.

Strikingly: they are not more accurate about men’s preferences compared to women's (who they engage less with)

I see sensible trends in accuracy across issues:

Politicians are more accurate on issues they engage more frequently with (but still closer to the random guess benchmark)

More fine-grained measures conclude the same.

See the paper for more on the correlates of accuracy!

Showing this information gap is not enough to establish that what politicians know matters for accountability & public good provision.

They might simply not care.

I next ask whether experimentally alleviating the gap affects how politicians act.

Sample: 653 local politicians.

EXP DESIGN:

⏺️Treatment: Info about what citizens prefer in binary trade-offs

⏺️Randomize citizen type (gender & partisanship)

⏺️Randomize T/C within politician

⏺️Outcome: What politicians recommend to their party leadership (I observe this because I partner with the party)

📢Information matters for politicians’ actions!

In control: politicians recommend the majority’s preferred policy (in binary trade-off) only 52.5% of the time.

In treatment: 7.6 pp. (14.5%) more likely to do so.

Effects (i) don't quickly dissipate (i) not driven by exp. demand

Next, I ask whether responsiveness varies by the type of politician who gets info and by the type of citizen whose preferences are provided

Politician type = directly vs. indirectly elected ➡️ imp. for institutional design

Citizen type = gender & partisanship ➡️ imp. for equity

📢How politicians are elected matters for responsiveness

I find that directly elected politicians are significantly more responsive than indirectly elected ones.

This suggests that politicians' incentives affect responsiveness ➡️ local governments should have direct elections.

📢Information is not used to favor co-partisans

Politicians are not more responsive to the preferences of their own party's supporters versus the electorate at large.

This shows that politicians do not use new information to discriminate in favor of their core supporters.

📢Politicians are more responsive to information about women’s preferences

This implies that in a context where politics is a largely male domain, returns to increased women's participation may be high.

Why are they more responsive to information about women’s preferences? 🤔

Under my model, this happens because they are less confident in their prior beliefs about women compared to men. When they receive a new signal about women, they place a higher weight on this signal and hence update their beliefs more.

I provide suggestive evidence for this channel using survey responses showing that politicians are thrice as likely to state that they know more about the preferences of men than women (+ consistent heterogeneities). I provide evidence inconsistent with other explanations.

But if politicians value this information, why aren’t they better informed in the first place?

The reasons are gendered:

(1) They believe they less new information about men
(2) social norms prevent them from acquiring more information about women (cue: my other work)

Does exposure to info affect politicians’ demand for more info?

If info has📉returns, exposure may📉demand.

📢Instead of decreasing demand, treatment increases politicians’ demand for information.

But, only on dimensions they pay less attention to under status-quo

⏺️What are the implications of these findings?

(1) What affects public good provision is a big Q for dev econ. I show: what politicians know about citizens matters!

(2) Even when women are excluded, policy may shift closer to their preferences if politicians are better informed

⏺️What are the implications of these findings?

(3) We think political contact = good for responsiveness. I show more contact (with men) can instead undermine representation.

(4) Information frictions matter in politics (as they do in labor, health, energy)

⏺️Policy recs:

(1) Better mechanisms for information flow from citizens to politicians—that include the underrepresented

(2) Better representation of underrepresented (new local govt. reform on this!)

(3) CSOs must act when parties don’t do this due to incentives/inertia

This paper has taken me to really exciting places (both geographically and intellectually), and SO many people from academia, politics, govt helped me answer these questions. Research is really a collective endeavor, and these acknowledgments only scratch the surface.

The full JMP is here: bit.ly/2KyR7vt

Check out more of my research on asadliaqat.com

e.g. my co-authored paper with @_sarahkhan, Ali C and Shandana KM on constraints to women’s turnout: bit.ly/35a9pLe

Thank you for reading🙏

I would💚 feedback!