Wednesday, February 25, 2009

Consumer confidence plummets, and so do expectations

The Conference Board released the results of its February Consumer Confidence survey. From the Washington Post Ticker: Consumer confidence in February hit another all-time low, as Americans struggled with layoffs and a recession.

It's bad out there, folks. To be sure, the February reading did set a record low since 1967, but the record is just one month old, as the January reading was the previous low. In fact, except for the November 2008 bump, each month has set a new record low since September 2008. The labor market is just brutal, and that is killing confidence. Consumers are getting gloomier and gloomier on an uncertain economic future.

Speaking of future, according to the Conference Board survey, consumers don't see light at the end of the economic tunnel for the next six months.

This chart, which illustrates consumer business expectations over the next six months, is particularly disconcerting for two reasons. First, the percentage of consumers surveyed that believe conditions will worsen surged by 9.4% to 40.5%. This is troublesome, since waning confidence in future conditions likely drives down current spending by increasing savings. Yes, this is a good thing over the long run, but it can further contract the macroeconomy hard in the near term.

Second, the number of consumers that claim conditions will not change in 6-months is falling quickly; or an increasing number of individuals are developing an opinion that the economy over next six months will corrode.

The percentage that believe that conditions will not change over the next six months is not listed, 50.8%, but it is simply 100% minus those that believe the economy will worsen, 40.5% in Feb., and those that believe the economy will improve, 8.7%. Notice how there is always a rather stable number of people that believe conditions will be the same over the next six months? Well that has changed abruptly. All the previously "same" people are falling into the "worsen" boat.

Overall, this is a terrible report. I imagine that confidence will increase when the fiscal stimulus money starts to roll out; but until then, there is really no reason to be positive right now.

Rebecca Wilder


  1. Hey Rebecca,

    Confidence is key. According to Dismal Scientist (Moody's, the Washington Post/ABC News poll saw a one-point improvement from last week, but it still remains "noticeably depressed."

    At what number will this survey be seen as "confident" or "back to normal?"


  2. 2000--- jan--- top
    2002--- oct--- bottom
    2006--- feb--- top
    2006--- oct--- bottom
    2007--- feb--- top
    2007--- apr--- bottom
    2007--- sep--- top

    These are the swings in the rates-of-change in legal reserves (the #'s are now flawed). They are essentially seasonal. If there is a bounce it will be in apr.

    One way to trade the seasonals is to go with or against the seasonal turn. I.e., if the market moves against the turn, short the market, if it goes with the trend, buy the market. You must trade on the seasonal date, e.g., may 5.

  3. The Consumer Confidence Survey of the Conference Board amounts to worthless rhetoric.

    Yet, year-after-year, many newly minted economists from Academia get trained to accept such useless propaganda.

    One fool-proof method exists to know definitively what consumers believe by what consumers do -- sales tax receipts relative to disposable personal income.

    As voluntary taxes, an increase in sales of taxable goods relative to after tax income shows that consumers have willingness to buy more luxurious things thus showing their confidence in their future forthcoming income.

    On the Bogus Consumer Confidence Survey

    First off, each state in the USA has local economic constraints owing to variant tax rules, population count, industries.

    Thus, the survey would need to have 50 surveys, one for each state.

    Secondly, all public opinion polling must capture the opinions of respondents to the same news on the same day at the same time.

    If polling happens over several days, all the poll taker has done is collect the opinions of the respondents, which cannot be projected to the universe of all adult Americans.

    In short, to obtain a margin of error of +/- 5% at the 95% confidence level, a poll would need 18,800 completed questionnaires (50 x 376) all completed on the same day at the same time.

    Since this is not the method of the Conference Board, the Conference Board reports bogus results each month.

  4. Smack, please tell us - how do you really feel?

    What about the sentiment survey?


  5. You know me by now Rebecca.

    I'm for truth, always, and helping others to come and enjoy it with me.

    We know humans by their acts, not by their disposition.

    Plenty of negative men and women abound. Yet, in spite of their beliefs of gloom and doom, their acts say otherwise.

    Betting on acts trumps betting on whimsical feelings, any day.

  6. Just look at how the market reacts to news. Consumers do the same thing from moment to moment, changing with the winds, so I don't have very much faith in "consumer sentiment". Have heard, though, that when it gets really, really bad, things have started to improve.

  7. Janie, you are right!

    This is why most public opinion polling yields bogus results.

    News changes beliefs.

    The lag time between completing enough questionnaires compared to learning of new news is too great.

    Also, quite often sample sizes and sample design prove to be wrong.

    Thus, any results cannot get projected to the defined universe of others.

    Polls end up being only the useless opinions of those polled, none of which can get projected to the wider populous.