Zombie Economics

Zombie Economics popped up on Scribd. I’d seen some bits of it on Crooked Timber so I thought I’d read it. I doubted I’d get much new big picture over having read a bunch (esp Krugman) during the financial crises of 2008, but I was hoping for some depth.

I was really disappointed.

It’s readable enough. And there’s lists of papers at the end of each chapter. But it didn’t feel very informative with the polemic being more rhetorical than substantive. Part of this, I’m sure, is the general reader audience. Consider this bit on privitization:

Most evidence on privatization comes from developing countries and is decidedly mixed. There are favorable cases, such as that of the steel industry in Brazil  where privatization turned loss-making and declining public enterprises into  profitable and growing private corporations. On the other side of the ledger, there  are cases like that of Russia where privatization was the occasion for wholesale  looting, allowing self-described democratic “reformers” to enrich themselves mas-  sively. Most cases fall between these two extremes, but the view that privatization  is always, or even mostly, beneficial is not supported by the evidence. 

Examining a number of actual privatizations in Australia, I found that the government made net fiscal gains in only two cases. In both cases, the sale took  place in a bubble atmosphere, with the result that the buyers subsequently resold  at a loss. Looking at cases where privatization was proposed, but did not go ahead,  the returns to government under continued public ownership clearly exceeded  the benefits they would have obtained from selling the assets. On balance, there  was a net fiscal loss from privatization in most cases. This was not offset by benefits to workers (who were mostly worse off) or consumers

If we look at footnote 7:

As discussed below, these were the privatization of the Victorian electricity  industry in the early 1990s and the second stage of the privatization of Telstra,  the former telecommunications monopoly, and also the dominant Internet service provider. In the Victorian case, the deregulation of the U.S. electricity industry had produced a group of cashed-up buyers, competing for a limited pool of assets. The Telstra sale took place during the dotcom boom. 

Oy! We’re in squishly land here. He examined “a number” of cases (how many? out of how many?) I mean, getting net fiscal gains in only two case is less impressive if there are only two cases in toto! [Ed note: Reading Quiggan’s response, I’m worried that this might give the impression that I think Quiggan’s being deceptive. I don’t think so!]

The first paragraph sets up two extremes and then says “most case fall between”! What?! This is supposed to refute the idea that privatization is always or “mostly” beneficial (to what? government finances? consumer choice?). Why not show at least some of the evidence? [Ed note: in line with the prior note, Quiggan says in the comment that it’s tricky to present and technical. I believe him, though I’d like to see an attempt. So this complaint just is, “It’s for a general audience.” I’d still have preferred a bit more caveating about how it’s tricky over his categorical tone.]

This sort of thing makes the book largely useless, I think. At least for me. There are some howlers, too:

Given the  state of the U.S. economy and the failure of the Obama administration to do much about it, the Republicans ought to be guaranteed of a sweeping victory in  the 2012 elections. In reality, the process of nominating an alternative candidate  has descended into farce, and the congressional Republicans have plumbed unheard of depths of (un)popularity. In large measure, this reflects the total sepa-  ration between ideology and reality now required of Republicans. 

Is the bolded sentence true? Or even justifiable? Remember that prima facie, being an incumbent is a strong place to be and Obama is super talented as a retail politician and he was extremely motivating for his base and he had incredible campaign structure. Is it obvious that the Republicans should have had a ginormous edge? I don’t think so. The way I’d adjudicate it is by looking at what various “fundamentals” models predicted.

And this is more of a PITA to dig out than I wanted for a quick blog post. Grr. But here’s two bits of evidence:

  1. A Nate Silver article from 2012 wherein he writes:
    “Although the model — which is distinct from the electoral map put together by The Times’s political desk — relies fairly heavily on polling, it also considers an index of national economic conditions. Right now, the polls and the economy are broadly in agreement with one another — both point toward an extremely tight race — although the economic risks to Mr. Obama are somewhat to the downside.

    Still, while the economic indicators suggest that the economy is growing sluggishly — at a below-average pace of about 2 percent growth per year — it is not yet in recession and incumbent presidents often receive the benefit of the doubt from voters. A favorable precedent for Mr. Obama is George W. Bush, who narrowly won re-election in 2004 under similar circumstances.”

    So the “state of the economy” was similar to another election where the incumbent was re-elected!

  2. The Moody’s economic trend based prediction model said that economic factors predict an Obama advantage:
    “In 2000, Moody’s Analytics began developing a model of the economic factors influencing American presidential elections that can be used to forecast the outcome. The 2008 version of the model proved accurate, predicting not only the election results that year but also the number of electoral votes received by the winner, Barack Obama. This year the election model has been updated and modified to take account of the recent recession, the deepest seen in many decades.Assuming the economy takes its most likely course between now and November, the model’s initial forecast calls for President Obama to win a second term. This prediction is tied to the Moody’s Analytics current baseline forecast for U.S. growth, which assumes that most states will continue to recover at slow to moderate speeds.”WHAT?!?!?!? Since Quiggin has given no detail about what the conditions were that were supposed to produce a strong Republican advantage, I can’t say it’s been refuted. But a big ole burden of proof shift has happened!

Now Quiggin is an economist, not a political scientist or election forecaster. His claim is exactly the sort of thing that lots of people are prone to make post facto. Indeed, it’s exactly a kind of zombie idea (cf Kerry as terrible candidate, Clinton as terrible in 2008, etc…these start with a claim that obvious these candidates had overwhelming advantages, then squandered them; the poly sci literature generally says that fundamentals dominate candidates or campaigns, esp in the modern era of polarisation).

And yet. It’s something he should have checked out. [Ed note: From his comment, I’m not sure that he did.] Fundamental models are hardly esoteric for people like him. It makes me nervous about the rest esp as I know less about those topics. The effort to check is high and I don’t have a good feel for what’s sensible and what’s fishy. So I, in principle, have to check it all.

And that could be fun, but it’s writing a whole other book!

[Ed Note: So, is this an exceedingly uncharitable review as Quiggan claims?

I don’t think so. I believe that the things that griped me are mostly things done for the sake of a general audience. So I am not the target. OTOH, I don’t know I would recommend it to a general audience because I worry about details like this and that it would do more net harm than good. I don’t feel that I get a toolbox for understanding (even at the lay level). And I don’t like even tangential strange claims which filter out.

I need to reread the Worldly Philosophers. I remember loving that in high school and feeling that I gained a lot of (lay) understanding. But maybe I wouldn’t like it today. Maybe I want to read textbooks.]

Advertisements

6 thoughts on “Zombie Economics

  1. This is an exceptionally uncharitable review. Obviously you didn’t like the book, but you’ve chosen to pick nits rather than explain why.

    On your first point,as you say, if i aimed at tricking my readers, I could have written ” two cases of success” when i’d only looked at three. I assumed readers would accept my implicit assurance the actual number was substantially more than two. As far as I know, you are the only person who has suggested otherwise.

    To repond more directly an exact count is tricky, because some enterprises like Telstra were privatised in stages, and because some privatisations covered multiple enterprises, but the number would be at least ten on a conservative count, and more than 20 if I counted each stage of each enterprise separately (electricity in most states and territories, Telstra in three stages, probably a dozen others ). Again, while i could have cited my studies in the further reading section, they are mostly fairly inaccessible, unlike the more theoretical papers I do cite.

    On your second point, it’s obvious that I’m using an implicit version of Silver’s model, in which the economy is the main driving force. We differ on the question: how bad was the economy in 2012? Silver’s measure which you endorse suggests, about the same as 2004. I say, a lot worse, based on employment/population ratios and median household income. In that case, as I wrote, the poor polling performance of the Republicans needs explanation. This is a point on which you have (AFAICT) zero expertise, and Silver has too few data points to discriminate between his measure and mine. This is a point on which reasonable people can differ, not a “howler” on my part. In any case, this is a marginal aside, obviously tangential to the main theme of the book.

    • Hi John! Thanks for commenting.

      It’s not quite a review, but a reaction. I focused on a couple of bits that stood out for me in my memory.

      I came in with very positive feelings and left with unhappy ones. Take that data point as you will.

      My flight’s about to take off so I’ll try to respond to the substantive points tomorrow.

      Again, thanks for commenting.

    • Ok, here goes.

      This is an exceptionally uncharitable review.

      I don’t think so. Your reading of it obviously isn’t very friendly overall, but your specific charge is:

      Obviously you didn’t like the book, but you’ve chosen to pick nits rather than explain why.

      I did explain why:

      But it didn’t feel very informative with the polemic being more rhetorical than substantive. Part of this, I’m sure, is the general reader audience.

      It makes me nervous about the rest esp as I know less about those topics. The effort to check is high and I don’t have a good feel for what’s sensible and what’s fishy. So I, in principle, have to check it all.

      So my complaint, roughly, is that you don’t show your work. I gave two places which jumped out at me as your not having shown your work where I knew a bit. And…you haven’t. Thus I have to check everything, even things I know much less about. That makes the book rather less useful to me.

      Thus, I’m not in your audience.

      I don’t see that I’ve failed to explain why I don’t like it or to qualify my dislike as “not being the right audience”. How this adds up to *exceptionally* uncharitable is mysterious to me.

      On your first point,as you say, if i aimed at tricking my readers,

      I certainly don’t mean to suggest that you were aiming at *tricking* your readers there. Just that *for me* I want the data. Hmm. Juxtaposed with the “howler” right after I can see why you might feel otherwise. I’ll try to clarify after this comment.

      And I agree that it’s tricky, but I’m the kind of reader whose interested in that and the other papers. So it doesn’t work so well *for me*.

      On your second point, it’s obvious that I’m using an implicit version of Silver’s model, in which the economy is the main driving force.

      You miss the point about what I found howlery. Fundamental models are standard. Almost all incorporate economic factors. You don’t get to make a categorical claim without reference to the fact that the relation between various economic factors and election outcomes is contested. At the very least you should point out that *your* fundamentals model differs from most (I believe; I could only quickly find two off hand, but I seem to recall that most were in agreement).

      I stand by this. It’s an error. Not that your model is wrong, necessarily, but you presented it as *fact* not as one model. (Interesting you ignore the Moody’s model!) So my answer to this:

      This is a point on which reasonable people can differ, not a “howler” on my part.

      Is that yes it’s a howler. It’s not a howler that refutes the rest of your book (as I point out!) but makes me uncertain about what to trust. Where I have less knowledge, I have to trust more. This line damaged that making the book less interesting for me.

      In any case, this is a marginal aside, obviously tangential to the main theme of the book.

      Yes, which is why I focused on how it made me less comfortable with the rest (not that the rest was wrong).

  2. Thanks for responding

    As regards the target audience, I wouldn’t have thought PUP and I could have made it any clearer that this was not aimed at an academic audience, in search of footnotes and careful qualifications. Didn’t the blood dripping down the front cover give you a hint? I have some sympathy with readers who complained that the book was heavier going and more academic than the cover led them to expect, but you are the only reviewer to suggest the opposite.

    More uncharitable reading. You now say “Not that your model is wrong, necessarily, but you presented it as *fact* not as one model.”, but maintain your claim that I’ve committed a howler. Observations
    1. If my model isn’t wrong, the fact that other models, published after I’d written the book, give different results is not a “howler”
    2. On your weaker criticism, only the most hostile reader would take a sentence like the one you quote to imply that anyone with a different view of the state of the economy as it affects voter behavior must be wrong. It was obviously a statement of my opinion, not an attempt to foreclose debate.
    3. Finally, given that (in my possible uncharitable reading) you suggest that I’m an ill-informed amateur, compared to your expert knowledge of political choice modelling, I’ll point out that I was estimating discrete choice models when Nate Silver was in elementary school* http://onlinelibrary.wiley.com/doi/10.1111/j.1477-9552.1986.tb01592.x/abstract Judging by your web page you seem to have an impressive track record in comp sci, but nothing that indicates a deep knowledge of this topic.

    * Not that I’m running him down. He’s very impressive and by far the best modeller writing on these subjects for the general public. To paraphrase your point, I don’t claim his model is necessarily wrong, but it isn’t the only model consistent with a fairly limited data set.

    To end on a positive note, we seem to be in agreement on lots of things, including a love of folk music. I’m sorry you didn’t like Zombie Economics, and hope that you might find more value in my journal articles.

    WordPress.com / Gravatar.com credentials can be used.

    • “As regards the target audience, I wouldn’t have thought PUP and I could have made it any clearer that this was not aimed at an academic audience, in search of footnotes and careful qualifications. ”

      It was clear. What wasn’t clear to me was that I wasn’t in that audience. I thought it was pretty clear from the post that I didn’t like it because I turned out not to be in the target audience.

      I think you’re still confused about what I found as a howler (hence your uncharitable reading). Not, I’m not an expert on election forecasting, just an interested layperson. But *as an interested layperson*, I’m aware of the plethora of “fundamentals” models that political scientists have been promulgating for decades. There are three sorts of predictive models I’m aware of, 1) “fundamentals” based which look at things like incumbancy, presidential approval, and perceptions of the economy, 2) poll aggregation and extrapolation, and 3) prediction markets. Hmm. I see PollyVote breaks out “index” and “econometric” models:
      http://pollyvote.com/en/components/index-models/

      What I was looking at to show that what you wrote :

      “Given the state of the U.S. economy and the failure of the Obama administration to do much about it, the Republicans ought to be guaranteed of a sweeping victory in the 2012 elections. I”

      is a howler was something like this:

      http://pollyvote.com/en/components/econometric-models/

      which lists all the forecasts for 17 econometric models (there’s also 5 index models including the famous Keys to the white house:
      http://pollyvote.com/en/components/index-models/keys-to-the-white-house/

      But I couldn’t find the 2012 ones easily alas).

      You just can’t say categorically state what you stated. The relation between the economy and the election is complex at best, and the two models I did find both gave the edge to Obama.

      And this is where your writing for a general audience makes this worse. Most people *aren’t* aware of this, so they’re inclined to accept it fairly uncritically. The idea that your categorical statement is going to be read as speculative or even one judgement amongst several seems…unlikely. It’s felt like a smooth part of your general polemic. A polemic about how ideological bias prevents people from seeing obvious contrary facts. I don’t see clear markers that this is more opiniony, sorry.

      I don’t at all suggest you are an ill informed amateur at anything, particularly in comparision to me (this is really weird on your part!). I feel that 1) you probably know about the plethora of fundamentals based models, 2) if not, you should, and 3) thus I should expect you to be clear that it’s not so obvious that given the economy, Republicans were strongly favored.

      Aaaargh, only *now* do I find the 2012 pollyvote discussion:

      https://www.researchgate.net/publication/265380457_Combined_Forecasts_of_the_2012_Election_The_PollyVote

      Which leads me to:

      https://www.cambridge.org/core/journals/ps-political-science-and-politics/issue/3740C69F1E135BA703A313A10D774712

      Which has a lot of the econometric models. The editors intro:

      https://www.cambridge.org/core/services/aop-cambridge-core/content/view/D2F44137CCF18C6BF2391A32AE5281A4/S1049096512000856a.pdf/forecasting_the_2012_american_national_elections.pdf

      table 2 lists the specific models and their predictions. (Finally!)
      “”””Five of the
      forecasts predict that Barack Obama will receive a popular
      vote plurality (though three of these are on the cusp of pre-
      dicting a toss up), five predict a popular vote plurality for
      Mitt Romney, and three regard the election as a toss up. The
      forecasts range from 53.8% of the vote for Obama to 53.1% for
      Romney. The median forecast is 50.6% of the two-party pop-
      ular vote for Barack Obama.””””

      Now maybe the economy centric ones are more pessimistic? I don’t know.

      In any case, I don’t see your claim, without some qualification, is reasonable. The models are split. So maybe your predicting is justifiable (I wasn’t sure when I wrote the post and the two models I found suggested not), but I’d like to see the justification. Your linked to paper isn’t an election forecast…so, I don’t quite get the relevance. I’m sure you are great at that stuff…I mean that sincerely. If you have an econometric presidential prediction model in your pocket, then I would suggest citing it in the book.)

      If you had said, “If we look at the range of econometric models for presidential forecasting, we see that the ones with a heavy economic component generally predicted an Obama loss [er…if this turns out to be true]…” I’d have no quarrel. If you had written, “The Obama economy was a real drag on his reelection campaign and, from what we can tell, it was only incumancy, Obama’s campaigning talent, huge turnout from the Obama coalition, plus relatively weak campaigning by Romney to overcome it” I’d still be a bit quibbly but that lets people who aren’t familiar (even as laypeople) about presidential election factors to realise what’s going on.

      Thanks for the positive note. You might see from this later post:

      https://bparsia.wordpress.com/2018/01/11/a-bad-paper-about-federalism/

      that I’m rather grump these days about such nits.

      I shall try to read some of you pertinent journal articles and discuss them here.

  3. Pingback: Intellectual Grouchiness | Mitigated Frenzy

Comments are closed.