Modelling as agit-prop

The latest issue of Agenda has just been published online. It includes a series of papers that investigate the misuse or abuse of modelling for policy purposes. Each of them is well worth reading (and yes, by pure conicidence, I have a paper in that collection).

If I have to pick out a favourite (and not my own) it would be John Humphrey’s analysis. He estimates that rather than ‘creating’ over 200,000 jobs the stimulus package “has resulted in the loss of over 30 000 jobs.”

Then there is Henry Ergas’ introduction to the papers:

The greater the difficulty consumers have in distinguishing the quality of modelling, the greater is the risk of ‘junk modelling’ dominating. Increasing the risk is the fact that poor-quality studies can masquerade as good-quality models by replicating their ‘look and feel’: for instance, through sheer size and complexity (as in the reports on the NBN discussed in this forum by Kevin Morgan), technical sophistication (as in the climate-change modelling discussed by Ergas and Robson), opacity and impenetrability (as in the modelling of royalties discussed by Pincus, of company tax forecasts discussed by Davidson, and of the stimulus discussed by Humphreys) or even simply by generating seemingly very large and highly newsworthy numbers (as in the modelling of the costs of congestion discussed by Harrison) — and, most often, by a combination of all of these. The fact that key interpretative economic concepts are often misunderstood and misapplied (as in the modelling of ATM regulation discussed by Green) then allows questionable results to be parlayed into effective advocacy.

Anyone interested in policy and the modelling that often underpins policy proposals should spend some time reading this issue of Agenda.

This entry was posted in Uncategorized. Bookmark the permalink.

60 Responses to Modelling as agit-prop

  1. Skuter

    Ergas’ summary as quoted above almost reads as a bureaucratic/political ‘how to’ manual for public deception…

  2. Ash

    The other side of the story is of course data quality. i.e., does anyone believe the official unemployment data is reflecting what is actually going on?

  3. jasbo

    Glad this is on the radar. Anyone who has been involved in social policy research can attest to the research-as-advocacy problem.

    Ask yourself this question: how long has it been that a known researcher, or established research centre, has published anything that wasn’t predictable from their prior positions? That in itself raises questions – on all sides of the policy debate.

  4. Econocrat

    From The Treasury’s Non-modelling of the Stimulus

    The ignorance of open-economy macroeconomics suggests that Treasury has neglected much of the advances made in macroeconomics over recent decades, and its strange assumptions on domestic crowding out and private savings response show that it has forgotten much of its own research.


    When I worked at Treasury, it was clear that they had not advanced from 1960s mixed-economy thinking. Every model I learnt at uni was unknown to them, and macroeconomics was basically Y = C + I + G + (X − M) with a bit of eyeball econometrics (charting) thrown in.

    A disgrace.

  5. Ash

    I am not sure if it is such a problem that research, at times, is predictable. In my opinion, the real problem is that research/model output is often presented as the “whole truth and nothing but the truth”. Further, the propaganda associated with findings nearly always crowds out critical assessment. Not to mention a total lack of transparency of what the real and important limitations of the findings are – think Treasury projections.

    I think we would all have a better understanding if we recognised that no model is perfect, or ever will be! Assuming they are reasonably robust however, they do can provide valuable insights into the (in)effectiveness of Government policy.

  6. Dr Faustus

    Assuming they are reasonably robust however, they do can provide valuable insights into the (in)effectiveness of Government policy.

    That there is a hell of a big assumption, however your general point is a good one. The Ergas and Robson paper, Modelling as Agit-prop, is an example.

    The Treasury modelling for the Carbon Tax suffers from an impressive variety of deficiencies, including a fundamental howler: the comparison scenarios both assume the whole world implements a uniform and efficient carbon pricing system by 2020. Only Australia opts in, or out.

    The modelling is a complete nonsense; a comprehensively terrible piece of work with absolutely no objective connection to its external realities. And it provides a valuable insight into the effectiveness of Government policy.

  7. Poor Old Rafe

    A nice piece from a sophisticated scientist that was written in the context of agricultural research but touching on many of the ways that economic models have gone wrong lately. Twenty years ago he was concerned with the disconnect between the views of politicians and the world of scientific research.

    There is another source of misunderstanding of the scientific community. A significant, well-run research project may require 5 years or more, a period often longer than the tenure of politicians, CEOs, and bureaucrats and, indeed, the duration of “ideas in good currency” in society (Schon 1971). Failure to grasp this can lead to a simplified view of accountability, the dangers of which must be recognized.

    He retired in the early 1990s and was killed by a tram in Amsterdam before he lived long enough to see the research world was completely overtaken by political agendas. But he saw managerialism coming, courtesy of the Harvard Business school.

    The consequences for science of instituting managerialism are saddening to behold. Unnerved scientific administrators, cowed by their management consultants, dismember scientific teams of world stature and set up systems of line management. With rare exceptions, the top rungs are occupied by failed scientists, by flawed scientists who have abandoned understanding in favor of power, and by gray bureaucrats blissful in their ignorance of what science is about.

    That came on the back of “scientific Marxism” which he noted had triumphed in the west even as it failed in the east.

    Beyond the professionals sit the decisionmakers. Understandably they are enthusiastic about models. Like the rest of us, they enjoy good news, and the news the messengers bring is that models, sanctified by the authority of the computer, will solve their problems. Decisionmakers can thus join the throng of those ducking personal responsibility: their decisions are forced upon them by the pronouncements of the computer. Conversely, it is not unknown for an unscrupulous decision-maker to seek out models and modelers that give the answers he wants. This, of course, requires modelers skilled in adjusting the model to yield the desired result; there is no place for the unpredictable output of the less skilled.

    Perish the thought of producing a result that the politicians don’t like.

  8. Poor Old Rafe

    There is a paper defending the use of macro economic models in the current CIS journal Policy. Any comments from the resident boffins if they have seen the piece?

  9. cohenite

    technical sophistication (as in the climate-change modelling discussed by Ergas and Robson),

    Is there an example of technical sophistification in AGW; that would be unique from my reading.

  10. Jim Rose

    one of the first things I learnt in the public service policy analysis is you can write a number on the back of an envelope after a long liquid lunch and it will go far.

  11. sdfc

    Yes the economics of a goods market in equilibrium, an exogenous money supply, an endogenous interest rate and perfect capital mobility. Where customers are just lining up to buy all the tellies we make, if only the exchange rate was lower and unemployment was higher.

  12. Cold-Hands

    one of the first things I learnt in the public service policy analysis is you can write a number on the back of an envelope after a long liquid lunch and it will go far.

    This still holds true in the Rudd/Gillard regime- NBN anyone?

  13. blogstrop

    Has anybody been modelling how long it will take to pay down the latest 200bn debt? When the conservatives got back into power in 1996 it was (gasp!) only 90bn, but that was after the outgoing mob said (in their last budget) no worries, a mere 10bn debt.

  14. Gab

    Is that so, Blogstrop? Labor lied and said the debt was $10 billion when actually was $90 billion? Holy pesos, imagine what the current debt is then?!

  15. Rabz

    Holy pesos, imagine what the current debt is then?!

    With respect to blogstrop, I thought the current nominal debt figure was at least $250 billion…

  16. .

    Yes the economics of a goods market in equilibrium, an exogenous money supply, an endogenous interest rate and perfect capital mobility. Where customers are just lining up to buy all the tellies we make, if only the exchange rate was lower and unemployment was higher.

    What a load of shit. You don’t even understand what you criticise. Keynes has atrophied your brain.

  17. sdfc

    All those points are true. The exogenous interest rate and exogenous money supply are a hoot. Exactly back to front.

  18. Rabz

    “scientific Marxism”

    WTF? Peddled by “marxist intellectuals”, no doubt.

  19. sdfc

    Make that endogenous interest rate.

  20. .

    No you are lying. Classical economics never claimed to use higher unemployment to end disequilibrium.

    Price rigidities are caused by regulation.

  21. .

    Make that endogenous interest rate.

    Entirely true. The empirical evidence is that M3 growth causes M1 growth.

  22. sdfc

    Don’t get excited Dot. My comment about unemploymnet being higher was a flipant comment. About the rise in unemploymnet that is likely to occur while we are awaiting the trade bonanza from our higher exports. Or from the factories that are going to spring up in the name of import substitution as a result of the lower exchange rate.

    I don’t think letting unemployment rise is an option when household debt to disposable income is sitting around 150%.

  23. sdfc

    Excuse the typos, had a couple of beers at lunch, had a couple more at home.

  24. JC

    The empirical evidence is that M3 growth causes M1 growth.

    So M3 which is defined as:

    That’s basically all of the money supply.

    M1 is defined as:

    notes coins in the hands of the public, Cheque accounts and what Americans refer to as NOW accounts, ie. Negotiable order of withdrawal ( I presume that’s something like a term deposit.

    So yea, a rise in M3 would have to be seen in a rise in M1.

  25. .

    Or from the factories that are going to spring up in the name of import substitution as a result of the lower exchange rate.

    …and yes industry policy and fiscal stimulus is going to react to changes in commodity prices?

    Lulz.

  26. JC

    Excuse the typos, had a couple of beers at lunch, had a couple more at home.

    You seem to have a problem with drinking, SFDC as every time you come on here you make a reference to this nasty habit of yours. Have you considered AA?

    I don’t think letting unemployment rise is an option when household debt to disposable income is sitting around 150%.

    And what exactly do you suggest we do when the country is coming off a comod boom and the fucking idiots in the government with the lying slapper being the main culrpit giving us a minimum wage rate which is the highest in the world.

    Look, Festus, I’ve told you this before. You should be concerned with aggregate income , not maintaining Joe Blog’s high and unsustainable wage rate.

  27. .

    You seem to have a problem with drinking, SFDC as every time you come on here you make a reference to this nasty habit of yours. Have you considered AA?

    Only when he otherwise would have to admit failure or concede defeat. “It was the demon drink wot dun it. Not me stupid ideas about economics.”

  28. JC

    and yes industry policy and fiscal stimulus is going to react to changes in commodity prices?

    Ah yea, the smart labor force theory senior big wigs in the cabinet keep talking about as a way to improve da productivity.

    I think we’re actually fucked. Very high regulated labor rates, a commod boom coming off the boil, a cabinet made up of ex sheletered workshop veterans and very low productivity.

    The leftards keep saying that we have been experiencing low productivity gains since 2004(ish). They say this to reference the point that it happened during Howard’s time too. However we were experiencing gains in participation rates during that time when the marginal worker was getting a job, so it was “good” low productivity then.

    The troubling point in my mind is that we are seeing lower participation rates and productivity is basically at a standstill now. This is really bad shit in my opinion.

  29. JC

    Dot, He’s always drunk. SDFC has a serious drinking problem in the same way fat boy has a food addiction. SDFC needs to go to AA and Fat Boy to the fat farm… sooner later than later.

  30. sdfc

    Nancy boy gets hysterical when man has a couple of beers. Rants about Gillard, demands lower wages in an economy where household debt is 150% of disposable income is 150% Funny as shit.

    Pathetic Dot.

  31. sdfc

    Why are you so scared of Monty and I JC?

  32. .

    Skilled worker shortage?

    What numpties. The unemployed cannot afford to pay for training to obtain what is now legislatively required occupational licensing for previously unskilled jobs, or jobs where on the job training provided semi skilled and skilled workers.

    We can attribute this to the union movement and their long march through human resources.

  33. .

    Nancy boy gets hysterical when man has a couple of beers. Rants about Gillard, demands lower wages in an economy where household debt is 150% of disposable income is 150% Funny as shit.

    Pathetic Dot.

    ???

    I don’t mind if you drink, just don’t become an unhinged big mary.

    You are a liar. Keynes also demanded lower real wages.

    Idiot.

  34. JC

    Rants about Gillard,

    Rant about the lying Slapper? Binge boy, Immentioned her in reference to the current labor market as she was the one instrumental in setting it up. Recall?

    That’s not a rant, you idiot. I’m stating a fact.

    demands lower wages in an economy where household debt is 150% of disposable income is 150% Funny as shit.

    I demand? Doofus if the commod boom comes off the boil we’re fucked with these regulated labor markets and the highest in the world minimum wage rate.

    I’ll give you a micro example. I’m currently managing a small family business getting it ready to sell after the holidays. The wage rate is basically legit 22 bucks an hour on average on the books. I was talking to a supplier and he suggested we were crazy paying that sort of wage rate. He hires off the books, straight off the boat Frogs, Italians, Greeks, Germans, Limey backpackers off some website called Gummtree for 10 to 12 bucks and hour.

    The real market wage is around that level it seems.

  35. sdfc

    Exactly Dot. There is much less pain in lowering real wages through inflation than outright declines in nominal income. If nominal income falls when there are high levels of debt you are basically fucked.

  36. JC

    Why are you so scared of Monty and I JC?

    Lol.. It’s the sturdy intelligence you show when you’re drunk and when he’s had a Mr. Creosote moment.

    You both ought to realize that gluttony is one of the worst 7 deadly sins.

  37. sdfc

    Of course you’re chicken shit. That’s why you resort to the twelve year old girl name calling. You’re a pansy afraid of the mean men. Weak as piss.

  38. sdfc

    What makes me laugh is you crap on about NGDP but don’t even know what it means. You’re a dimwit.

  39. JC

    Of course you’re chicken shit. That’s why you resort to the twelve year old girl name calling. You’re a pansy afraid of the mean men. Weak as piss.

    I don’t understand your anger SDFC. Fat Boy needs to get off the calories and I’m trying to help him.

    Shooting the messenger much?

    Look doofus, nothing would please me more if one day, Monster would show up here and tell us he’s dropped 80 kilos and looking to shed more.

  40. JC

    What makes me laugh is you crap on about NGDP but don’t even know what it means. You’re a dimwit.

    How can you say that?

  41. .

    I’m not advocating “falling wages”, sfdc. Keynes did.

    Prices should be flexible. They (mostly) are. Keynes argued only semi successfully about wages. This is irrelevant if wages are not centrally arbitrated. This view of incomes is very antiquated. Firms stopped paying bonuses, cut hours, prices fell and there was monetary and foreign prices accommodation. The crisis was averted because the labour market and external balances were flexible.

    John Humphreys has shown empirically the fiscal stimulus destroyed jobs. What we know off hand is that the stimulus saw capital formation slow as it was implemented. Each “job saved” cost many times over the annual average income. It simply could not have worked. You do realise the Keynesian model isn’t applicable to open market economies, do you?

    When incomes fall, prices fall anyway, unless all supply is permanently in the short run.

    If we have even 160% private debt to GDP and a work force of over 10 million, this is less leverage than a conservative mortgage on average. It is still many times less than an asset price deflated national stock of wealth.

    It is no big deal. Why you even bother bringing it up is a mystery.

  42. sdfc

    You’re a deadset goose. What grown man resorts to calling people fat as an argument. You embarrass yourself.

  43. sdfc

    We ahve evidence of what happens when wages fall in a high debt economy from the experience of the early 30s.

    John Humphries relies on the unrealistic assumptions I posted above.

  44. JC

    You’re a deadset goose.

    Name calling now, are we? That’s very hurtful SDFC.

    What grown man resorts to calling people fat as an argument.

    I don’t. I always explain why he’s wrong. The fat part is added on to make him aware of the problems associated with his current lifestyle.

    Have you seen him in the pics he’s put up of himself on public display?

    You embarrass yourself.

    Now you’re being silly.

  45. .

    John Humphries relies on the unrealistic assumptions I posted above.

    He relies on empirical evidence.

    Suck it up.

  46. .

    John Humphries relies on the unrealistic assumptions I posted above.

    I have shown they are not unrealistic, Keynes relies on all incomes being centrally set wages and all supply functions being permanently short run, but also your world view on money has been empirically disproven.

    Give it up sdfc. You are being hammered here.

  47. JC

    We ahve evidence of what happens when wages fall in a high debt economy from the experience of the early 30s.

    No dickhead. You’re looking at the symptom rather than the cause. It’s like you’re treating a fever brought on by lymphoma with Penadol.

    The Fed cheered on by the hard currency brigade caused the drop in wages.

    John Humphries relies on the unrealistic assumptions I posted above.

    He does not. You’re lying.

  48. sdfc

    No Dot Keynes did not rely on wages being centrally set. Keynes was in favour of lowering real wages through inflation. Largely because wages weren’t centrally set.

    I have described the assumptions accurately Dot. Humphries uses simple Mundell Fleming analysis. Its assumptions are incorrect.

    You’re either a deflationist or you’re not JC. Make up your mind.

  49. JC

    You’re either a deflationist or you’re not JC. Make up your mind.

    Doofus, if the free market wage is 10 to 12 bucks an hour (in other words that’s what the grey market is purporting, there is no fucking way in hell that inflation would cause the nominal wage rate to fall to a real level of the current grey rate of 10 bucks an hour in a short period of time. The inflation rate would be so huge the RBA would not embark on it.

    Here’s the point that you keep missing. The thing that should concern you and the only thing is aggregate income, not labor rates.

  50. .

    No Dot Keynes did not rely on wages being centrally set. Keynes was in favour of lowering real wages through inflation. Largely because wages weren’t centrally set

    Yes he did. He was writing about 1930s Britain.

    I have described the assumptions accurately Dot. Humphries uses simple Mundell Fleming analysis. Its assumptions are incorrect.

    Wrong, I have destroyed your assumptions, you simply assert to say they are wrong relies on other incorrect assumptions which supports you, this is a circular argument, and the empirical evidence shows that you are incorrect.

  51. Jim Rose

    dot, keynesian marcoeconomics no longer relies on sticky wages. menu costs and sticky prices are the new crutch.

    sticky wages would mean real wages would rise in a recession making such times a pretty good time for most people holding stable jobs.

  52. .

    sdfc is a 1936 keyensian.

    Menu costs and sticky prices are irrelevant to actual changes in output.

    Real estate prices fell 40% in the GFC.

    If Keynesians think this isn’t enough, I fear what they believe is “enough”.

  53. sdfc

    Jim

    I’m not a new Keynesian so I don’t what you are talking about.

    Dot

    Keynes thought that workers would resist wage cuts because their income would fall relative to workers who had not had their wages cut. Reducing real wages via inflation was a solution.

  54. .

    Keynes thought that workers would resist wage cuts because their income would fall relative to workers who had not had their wages cut. Reducing real wages via inflation was a solution.

    Except they don’t.

    People had hours cut in the GFC and they understood why.

    There are/were in built institutional flexibilities Keynes assumed couldn’t exist.

  55. JC

    SDFC

    Raising inflation is not a solution at all. People who can, like bond markets will simply adjust for it. NGDP is the solution.

    Please stop the idiocy and the lying. Of course you;re a Keynesian zombie.

  56. sdfc

    Dot

    Workers resist wage cuts. It’s silly to argue otherwise. It does not mean that wages sometimes do not fall.

    JC

    Higher NGDP growth inplies higher inflation.

  57. JC

    Higher NGDP growth inplies higher inflation.

    It’s not the same thing. As Sumner once berated Brad de Short… “a thousand times no”. Think about it and if you’re still stuck come back to me and I’ll try to help you out.

  58. sdfc

    JC

    You can’t have two targets with one policy. To get NGDP growth back to target requires higher inflation.

  59. JC

    WTF are you talking about. There would be no inflation targeting in NGDP targeting.

    You’re making shit up as you go along.

  60. .

    Workers resist wage cuts. It’s silly to argue otherwise. It does not mean that wages sometimes do not fall.

    No, it is not. What “resistance” did we see in 2008-2009?

    You really are making shit up as you go along, this is beneath you.

    Higher NGDP growth inplies higher inflation.

    What a load of bullshit mate. You clearly don’t understand NGDP targeting.

Comments are closed.