Tuesday, May 30, 2017

A survey

My Hoover colleague David Brady put together this lovely graph, using the Gallup poll results from their standard question, "do you think the economy is getting better?"

It has many interesting interpretations -- add your own in comments.

A context that spiked my interest is the use of survey data for expectations in economics. We frequently use such surveys to establish what expectations are, or to question whether expectations are rational.  That partisan feeling so dramatically affects people's views of the economy is  interesting for the interpretation of such survey results.

Most economists don't think that presidential elections have all that much to do with the state of the economy, at least within the bounds of what traditional Republicans and Democrats actually end up doing when in power. The people, apparently, believe quite differently, or at least this time.


  1. if the Republicans don't follow through it will be interesting to see what that graph does-and what the midterms like. Some of the Republicans are acting like they forgot why they were given control. Repeal and Replace Obamacare and cut spending and taxes....

    1. pointsnfigures:

      But why did Trump blow the smithereens out of other GOP presidential candidates? It wasn't even close. Remember Jeb Bush getting 1.2% of the vote? This was Bush, the very embodiment of an establishment GOP'er. Crushed!

      I think it is that Trump was sounding themes that would tighten US labor markets: Restricting imports and illegal immigration. Getting the economy moving again. Trump also blasted companies for moving jobs offshore and said he wanted to cut taxes on the middle class.

      That is far different message that the GOP-estalishment tune, which is that yawning trade deficits are good, and there is nothing that can be done about illegal immigration, and US companies are right to move jobs offshore and yes cut taxes on rich people.

      (Also Trump said he would get out of Iraq and Afghanistan, often to rousing cheers.)

      Trump has failed in follow-through, and seems uncertain how to proceed. He is a big disappointment so far (although that is probably not a surprise).

      If the Donks can pick up the ball they have a shot at crushing the GOP-estabishment like Trump did.

      Talk about jobs, jobs, jobs, and tighter labor markets, and stop talking about feminists or activists who see sexism and racism under every rock.

      Side question: we have a central bank that targets keeping at least 1.4 people unemployed and looking for work for every job opening. Do you think the Donks could make hay with that?

    2. You raised such a good point John! In our valuation cooures- consumer sentiment and investor sentiment index are used as a forward looking indicators and yet these very indictors individuals and businesses use how to make investment decision in the immediate future which then reflects in the coming months index survey results. It does raise a question whether they are really forward looking indicators.

  2. This graph is one more piece of evidence that it makes sense to ignore what the Dems and Republicans have to say (because every question, to them, seems to be a Rorschach test) and just focus on the Independents. In this case, the Independents seem to perfectly chart the Trump trade.

  3. The graph was an awesome idea! Bravo to David Brady.

    My only (very mild) critique is that the chart needs to go back further in time. Right now this can be labeled "very interesting." If it went back to, say, the 1960's then it would become "informative" and perhaps even "useful" (depending what the data showed).

  4. Perhaps surveying voters' views is a tricky business. Like politicians themselves, respondents are not answering the original question about the economy but rather whether they think they will be better off themselves. If that were true, the chart could actually be revealing expectations about changes in tax policies.

  5. But do people's perception and predictions of inflation vary by party? I dont know what the term the economy means to people, I don't even know what average inflation means to people.

  6. The wisdom of crowds is based on offsetting biases, so the average expectation can still be pretty accurate even if both Dems and Reps are clearly biased.

  7. This is indeed an interesting graph. What it reveals is that people are actually choosing to give an answer to a different question, "Will the economy perform better under my lot than their lot?". This kind of response is quite usual in these types of surveys: in the ones I know most about (Australia and New Zealand equivalents), we have learned to discount the overall "business confidence" answers (which get contaminated by things like politics) and instead focus on "own activity" responses, which give a better indication of what people really feel about the economy in front of them.

  8. It shows that people may forget this passage: “Ask not what your country can do for you – ask what you can do for your country.”They may like big government, or people believe that only G (Y = C + I + G + NX) can help the economy “get better” ( = low unemployment or strong economic growth or…?). (But as your paper shows, G multiplier is not that big (or even negative) at ZLB!) Or people implicitly think that fiscal policy is more promising than monetary policy (such as QE or forward guidance).

  9. The economy has been improving pretty steadily since some time on 2009 or 2010. The stock market has been on an absolute tear since mid 2009. Throughout that period only approximately 25% of Republicans were willing to accept the truth that overall the economy was improving. This is what jumped out at me.

    The Democrats and the Independents seem to have been too negative but the Republicans have been down right delusional.

  10. You raised an excellent point John! We often treat consumer sentiment and investor sentiments indicies (both are based on previous month/fornight's survey data) as leading incators and yet we (individual investors and businesses) actually moderate our current 'economic' behaviour is based on previous survey results. The great point you raised (which I totally agree)- it must mean month to month survey data must have huge autocorrelations.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.