A Short History of Turning Points in Wisconsin

Ever since the Walker Administration ended the issuance of the Wisconsin Economic Outlook (last issue May 2015), those of us outside of the government have been a little in the dark when trying to assess state level economic conditions in a systematic manner. In a recent paper (updated version 8/1), Ryan LeCloux and I assess how various macroeconomic aggregates track the Wisconsin economy at higher-than-annual frequency. Investigating the behavior of these series yields the following graph.

Figure 1: Wisconsin Real GDP (blue), real personal income (red), coincident index (green), nonfarm payroll employment (black) and civilian employment (teal), all in logs, normalized to 2007Q4=0. Dark arrows (↓) denote peaks, light arrows (↑) denote troughs. NBER defined national recession dates shaded gray.

GDP peaks 2007Q3, a bit before nonfarm payroll employment (2008Q1), and coincident index, personal income and civilian employment (2008Q2). The trough is identified at 2009Q1 by GDP and personal income, and at 2009Q4 by the coincident index, and 2010Q1 by both employment series. Hence, regarding turning points, there is a high degree of coherence between the various series, at least insofar as Wisconsin is concerned.

What jumps out from the figure is that the coincident index trends differently that GDP, despite the fact that the trend component of the coincident index is calibrated to match that of GDP. The series might differ in practice because we are plotting a fairly short span of GDP (the period for which we have quarterly state GDP).

In our paper, we estimate the sensitivity of real state GDP growth (measured at national prices) to coincident index growth at 0.29 (signficantly different from zero at the 5% level, using HAC robust standard errors). From the trough in GDP onward, the sensitivity of GDP growth to index growth is 0.28 (significant at 5%); from the trough in the coincident index, the sensitivity is not statistically different from zero, unless the constant term is constrained to zero (the null of zero on the constant can’t be rejected at the 10% level).

One thing to recall that these assessments are made with a mix of revised and unrevised data, in a way consistent with how the NBER Business Cycle Dating Committee makes decisions regarding turning points sometimes long after the turning point has occurred. In order to call turning points in real time, we’d need to be able to measure at even higher frequency than quarterly, or be able to assess the relationship between different vintages of quarterly state GDP and the other variables, or both. That’s work we’re contemplating for future research.

In the meantime, we can assess what the coincident index, nonfarm payroll and civilian employment numbers suggested back in September 2008…

Figure 2: Wisconsin coincident index (green), nonfarm payroll employment (black) and civilian employment (teal), in logs 2005M01=0, as of November 2008. Civilian employment calculated using seasonally adjusted unemployment rate and labor force. Source: ALFRED, and author’s calculations.

As the graph clearly shows, despite substantial revisions to these series since 2008, the data as of November 2008 indicated a slowdown and possible recession was underway. The one ambiguous reading comes from the household survey based civilian employment series, due to its high volatility. That argues for relying more on the payroll survey-based series, and indicators based on those series (like the coincident index). For more on the household series, see this post.

33 thoughts on “A Short History of Turning Points in Wisconsin

  1. pgl

    Our resident Trump cheerleaders keep saying how we will win in the Trump trade wars and this will lead to an era of free trade. Conservative economist John Cochrane has heard this claim and he really does not buy it. In fact he likens it to the excuses that led to the war to end all wars aka World War I, which of course did not end wars at all (unless you think World War II never occurred):


      1. Bruce Hall

        Menzie, in the climate science world employment would be called a “proxy” for economic activity.

  2. Moses Herzog

    Because Iowa isn’t far from Wisconsin, and my German heritage, I’ve always found Wisconsin a fascinating and very likable place. Even a place I would consider retiring if my personal resources allowed that. You would think since Germans are generally known as a literate culture, in the state composed largely of ethnic Germans they would put higher regard to tabulating economic data. I’m assuming that if there was a Democratic governor or Democratic candidate for Governor he would be consulting Menzie or his campus colleagues on proper state policy on a daily or at minimum weekly basis.

    Oh damn, I just realized I’m being naive again and thinking politicians actually cared. Consider it a brain lapse on my part.

  3. Moses Herzog

    Reading you and LeCloux’s paper and enjoying it. Have a small list of questions but am trying to control the curiosity until I have finished reading and semi-digesting the entire paper. This topic is fascinating to me on multiple levels. Not the least of which would be the fact the state I reside in does such a crappy job of “tracking” economic data inside its own borders. I suspect on purpose, as the legislators would prefer the electorate to be running around blind to reality.

  4. Moses Herzog

    OK, I got my questions. I’m probably forgetting a couple, but these are the ones I got jotted down
    1—Why does it take so long to publish the QCEW numbers?? (is that because they are only used for revisory purposes??)

    2—As regards the research that you are going to do, but haven’t done yet: Were you and LeCloux trying to “come up with” or use more useful auxiliary variables in your research (for U.S. state governments to use)??

    3—Would it make sense (i.e be beneficial) to do “Keynesian type” GDP analysis at the state level??

    4—Would it be costly to get/acquire consumption data on a quarterly basis at the state level, if so, WHO would do that??

    5—Why is Delaware an outlier as it relates to the coincident indicators?? (Because national economy bears heavier influence on a small state??)

    6—Were you “burning the midnight oil” and very fatigued when you wrote the last sentence on page 23 (hint hint hint)???

    1. Menzie Chinn Post author

      Moses Herzog:
      1. QCEW is a census, not a survey. That means every firm that is covered has to be contacted.
      2. My motivation for the research is curiosity-driven: how to forecast economic activity at the state level. There is research, but it’s dispersed amongst researchers that do not communicate much with my community of academic forecasters, so I’d like to see if there’s scope for a profitable welfare-enhancing exchange.
      3. By Keynesian, I assume you mean a decomposition into spending components of consumption, investment, etc. I think yes, but I think it would be hard to do at the state level just because there so much cross (state) border spending.
      4. Consumption at the quarterly frequency could be estimated; I’m sure it’s just a matter of cost.
      5. Delaware has a large finance component, which breaks the link between (say) employment and output. Flora mentions this in his powerpoint presentation (cited).
      6. Probably yes, but it’s possible that it’s our inability to use the track-changes facility in Word competently.

      1. Moses Herzog

        Thanks for taking the time to answer. No joke, I would think if someone came up with a software package where you plug in these numbers and it helps with state level economic strategy, some “industrious economists” could uncover “a profitable welfare enhancing exchange”. If I had your energy and intellect, certainly that would be rolling around in my mind from time to time.

        Of course some governors could care less:

        Other state governors, like say a Schwarzenegger or a Jerry Brown, where BILLIONS are on the line, would be willing to pay a pretty penny on this. Of course maybe, in a way, the math and equations already exist on this stuff (??), but the question of how to get the accurate “inputs” into the equations is the confounding question. Let’s say you had 7–8 auxiliary variables that were cheap to access that could give you a very nearly accurate number on a coincidence indicator that was very expensive to account for, but important in government decision making. I would think a software package or “proprietary math equation” that could pop out that number would be the goose that laid the golden egg.

      2. spencer

        I suspect that Delaware is distorted partially because it is the state of incorporation for such a large share of corporate America. It must sustain a very large number of corporate lawyers in Delaware through thick and thin.

  5. Alan Goldhammer

    Of course Governor Walker failed on one of his central campaign promises of 250K new private sector jobs in his first term as governor. He’s still about 60K short according to statistics I have seen. Is there no accountability??

    1. Ed Hanson


      If you decide you want to see a lack of jobs than by all means keep looking at statistics you have seen. Governor Walker never defined what job parameters was to be used to measure performance. Menzie has chosen as well regarded but somewhat narrow measure of job growth. As, obviously, have you. I will give you a broad measure, statically accurate and useful, which shows Wisconsin reached the 250,000 mark way back in 2015.


      Jan. 2011 All Employees: Total Nonfarm 2,676,600
      Oct. 2015 All Employees: Total Nonfarm 2,929,600

      That is 253,000 more employees

      But actually the continued discussion of the 250,000 job promise is over. Over, not by appeal to statistics, but by the only realistic political measure. The voters of Wisconsin were satisfied with Governor Walkers efforts to revitalize the State, and showed that by his reelection.


      1. pgl

        A couple of details here:

        “Not Seasonally Adjusted”. I trust you realize what that means.

        Also note that in late 2007, this series had passed 2.9 million reported jobs. So job growth over the last decade has been incredibly small. Recovery from the bottom of a deep recession is not exactly long-term growth.

        But do continue to spin for Governor Walker since you note he “never defined what job parameters was to be used to measure performance”.

        1. pgl

          FRED provides a seasonally adjusted version of the same series that Ed Hanson gave us:


          Why Ed would insist on a series that is not seasonally adjusted is beyond comprehension. Lazy? Data mining? Or just plain stupid? I’ll let Ed choose among these alternatives.

      2. 2slugbaits

        Ed Hanson Not seasonally adjusted??? You flunked. BTW, Gov. Walker is trailing badly in his re-election bid. Maybe he needs to go to Moscow to look for another red headed girlfriend.

        1. pgl

          Not seasonally adjusted / seasonally adjusted. Either way – the net increase in employment over the last decade was a more 100 thousand not 250 thousand. Ed got an A+ from the National Review in cherry picking data!

        2. Ed Hanson

          To all

          As for statistics, there is nothing wrong with, in this case, using nonseasonal numbers. But of course, you missed the real point of my response, quite intentionally, I am sure.

          The 250,000 job number was answered in the Walker reelection for his second term. (No slug, it is not the third term election). The voters recognize the difference between the method of job growth compared to the political sound bite of results. Wisconsin voters liked the lower taxes, less regulation, and less power of the public employee unions. But that was all in the past. This election will hinge on different specifics.

          As for Walker being behind in the polls, it is mostly meaningless at this time. Only two polls are important. First the dem primary to select an actual nominee for governor, and then the second, the actual election.


          1. Menzie Chinn Post author

            Ed Hanson: There is nothing “wrong” with using NSA data. It’s just economists don’t usually use them. And even the Governor’s economists didn’t when defending the goal, except when they were doing *year-on-year* (e.g., December to December) calculations. (I am careful to specify economists, because one of the governor’s noneconomist hacks did try to do so, and was roundly castigated for it. You can find the episode documented in one of my year-end roundups).

          2. pgl

            You repeat this 250K number? You missed my point. That was a recovery from the Great Recession which every state has enjoyed – most more than Wisconsin. Try doing this from 2007 to today. Net employment growth was a mere 100K.

            I bet Stephen Moore admires your persistence at trumpeted distorted data.

          3. Ed Hanson


            I am in agreement, economist don’t usually use them. But this is a political matter, not economics. It has been brought to us over the years by political Menzie (using chosen economic statistics) not the economist Menzie (who would have analyzed economic results non-politically.

            But again, you are ignoring the important point, the political issue of the 250,000 jobs was resolved by the last election. The voters in that election clearly liked the policy implemented by The Walker administration or the electorate would have voted differently.

            And again, that is all in the past. This election, voters will make their voting decision of several factors, but right at the top will be ‘what have you done for me lately’ and ‘what will the opposition do to me in the future’.


          4. 2slugbaits

            Ed Hanson This election, voters will make their voting decision of several factors, but right at the top will be ‘what have you done for me lately’ and ‘what will the opposition do to me in the future’.

            Or equally likely, “Look at what your past policies ended up doing to me now. Fool me once, shame on you. Fool me twice, shame on me.” It’s the Kansas lesson. Voters are not too bright, but eventually even the dimmest voters begin to catch on.

          5. Ed Hanson


            you are in for a big surprise come election day, and for the very criteria you write of.

            I know you don’t realize it because you were a member of the protected government class, but the electorate still remembers the doldrums of the President Obama years and understand it takes time to correct for those years. It is tough on you, I know.


          6. noneconomist

            At one point–lasting about 8 years–Ed and the boys were horrified with “Trillion dollar deficits as far as the eye can see” Now, with a trillion dollar deficit looming next FY–and similar deficits on the horizon–they’ve happily adopted the Dick Cheney “…deficits don’t matter” mantra.
            That’s because they’re certain we’re going to have 8-9% GDP growth and deficits coming down (the debt too) like water pouring into a glass.
            We’ll just need to tweak capital gains taxes a bit more (index them to inflation) and the economic boom will hit the stratosphere. Trump said so. That means you can take it to the bank. Like having to show you ID when you buy groceries!

          7. 2slugbaits

            Ed Hanson you are in for a big surprise come election day

            Could be. The Russians are working hard for Walker. And judging by the smile on Walker’s face, he seemed to be enjoying the company of those red headed Russians as well.

          8. CoRev

            This comment thread has gotten funnier over time. Claim Walker never met his “jobs promise”! Ed, yes he did look here. Peanut gallery responses, not seasonally adjusted, economists don’t use that data, it was because of the recession recovery, blah, blah then change the subject to Walker’s election, and finally Russia, Russia, Russia.

            Nealy every subject gets this same treatment. Ad yet you can not see or more likely admit it is your own views and actions (comments) that has caused the “deplorables” to turn against your views.

            There is a conservative meme going around: Trump is a billionaire businessman, and President of the US. You? You just sit in front of your computer and rant”

  6. execon

    I would be interested in seeing cross-correlations that span maybe 1 or 2 quarter lags. Unfortunately, the quarterly GDP (by state) only allows us to analyze one recession episode (yet), but that single episode nicely demonstrates a possibility that labor market indicators are lagging and thus state “coincident” index is not actually coincident.

    In fact, this is true with the US quarterly GDI as well. (State GDP is mostly constructed in GDI-like manner but are scaled to national GDP plus/minus some adjustments.) For quite some time the labor income (namely “compensation of employees”) has been lagging behind the time-series movement of capital income components (e.g. corporate profits) inside the GDI account.

    Ironically, these arguably lagging labor market indicators are usually published way before profit/capital-side income data, which would make nowcasting of state GDP particularly difficult. To second Menzie Chinn’s answer to Moses Herzog’s question, unlike for US GDP, state-by-state expenditure data would have little explanatory power for state-level GDP movement without a reliable model/data that estimates interstate flows of goods and services. I personally think quarterly-or monthly- frequency estimates of such kind is practically infeasible, and I would be pleasantly shocked if I see a launch of new official statistics for annual cross-state economic flow in the US given the circumstances.

    1. Ed Hanson


      Nice one. “Wisconsin not even in top 5.” Its tough to be number 6, but close enough to 5.


      1. pgl

        In the Olympics, being #4 is no where. But in Ed’s hyper political BS, being #6 is almost as good as the gold!

        Glad you admitted today that all of what you do is about politics not economics. At least that is being honest per your intent!

    2. pgl

      New York is only 39? This survey says our health index and our lifestyle index sucks. Hey – things may not be perfect here but seriously?

      NY is near the bottom per “economic stability”. I guess this index does not value the level of income which is a lot higher here than in most states.

      Who came up with this index?

      1. Moses Herzog

        @ pgl
        According to 2015 numbers New York ranks 15th in per capita income, barely above North Dakota, That would mean over 1/4 of all other states rank above them. Those seem like old numbers but I can’t find more recent on per capita income.

        If you look at US. News and World Report numbers (With data compiled by McKinsey & Co) which has more factors included, they rank New York 35th out of 50, with would put New York in the bottom quartile of all 50 states. It appears to be a 2017 report but the years of the factors vary based on the most updated number available upon publication

        Not sure if that excludes NYC traffic, NYC arrogance, and NYC rudeness, but I’m assuming those 3 factors are excluded. pgl, You could phone the publisher to see if you can get them to include convenience of subway transportation if you feel that could boost the ranking.

        1. pgl

          Maybe it is because I live in New York City. The rest of this big state is basically farm country and given recent trends, their incomes may be hitting rock bottom.

          “convenience of subway transportation”? Giving how this is falling apart – my the inconvenience would lower our rating even more.

Comments are closed.