oops, we made a mistake

The budget papers have a great little box.

So, umm, those close links to Asia. What could they be? Oh, those minerals. The ones now being taxed to death?
Actually the government blames it on multipliers being too large.

This entry was posted in Uncategorized. Bookmark the permalink.

60 Responses to oops, we made a mistake

  1. asf

    By what measure was the US stimulus ‘small’. I thought it was about 7% of GDP? As far as I can tell South Korea’s was around 6% of GDP.

  2. Sinclair Davidson

    Those stimulus numbers look very dodgy – but I won’t be able to check the analysis until tomorrow.

  3. Rob W

    “By what measure was the US stimulus ’small’”

    By the measure of efficacy.

  4. Sinclair Davidson

    I can’t work out the base year for the stimulus size – in the OECD data I have the US is very large whereas here it isn’t. And I can’t find a source of actual GDP graoth for 2009. I can find a lot of forecasts but not the actual figures.

  5. asf

    Yeah they look funny to me too Sinc. In any case, just eyeballing the plot if you excluded Australia and South Korea I think it would weaken the relationship considerably.

  6. Sinclair Davidson

    Hmmm, maybe just taking out South Korea could do it. That’s just 11 observations.

  7. Peter Whiteford

    I suspect the size of the stimulus shown has something to do with the time period referred to in the IMF report. It is not clear whether they are referring to the 2009 calendar year or financial year and of course not all the stimulus spending in any country appears in the 2009 year, i.e. a lot would be in 2010 in some countries.

  8. Labor Outsider

    It is probably a timing and definitional issue. For a start, it is only discretionary stimulus, so excludes automatic stabilisers (which in most countries were responsible for the bulk of the deterioration in budget balances). And second, it is only 2009, so excludes discretionary measures that affect the 2010 calendar year. If you take a look at the past two OECD Economic Outlooks you will find a lot of information about how the timing and nature of stimulus measures varied across countries. My understanding is that Australia’s discretionary stimulus was indeed larger than most other OECD countries.

    If you take a look at the graph, it is basically a non-Japan asia story (I include Australia in this). Take out China, Korea and Australia and the relationship fades. I guess then you have to form a view on whether the Asian out-performance was related to their fiscal packages. FWIW, most China analysts believe that China’s stimulus measures contributed to their better than expected economic performance, even if they disagree about whether those measures were well targeted or wise.

  9. Butterfield, Bloomfield & Bishop

    I echo what LO just said.

    It does help to know what the OECD actually says in the data they give out.

  10. Pay close attention to what is actually being claimed here – from the Australian:

    http://www.theaustralian.com.au/in-depth/budget/wayne-swan-rides-tax-wave/story-e6frgd66-1225865282709

    *The government’s “decisive” intervention meant this year’s budget deficit would be $40.8bn — $16bn less than predicted a year ago. The performance, driven by what Mr Swan called the “commodity boom II”, would return the budget to surplus within three years — three years ahead of the expectations a year ago.*

    Is Swan claiming the stimulus caused the mining boom pt 2? It is also completely bizzare because the $57 bn deficit was predicted after the stimulus was underway. If they can’t forecast, why will this one be any more accurate?

    “And the jobless rate, forecast last year to peak at 8.5 per cent, had in fact peaked at 5.8 per cent and was forecast to go as low as 4.75 per cent by June 2012.”

    We all know this increase in unemployment was hidden in departures, discouraged workers and a very large rise in people not marginally attached to the labour force.

    From the comments:

    *Swan is basing his figures on a tax that has not been passed in the Senate and will probably be rightly rejected. *

    …but also the Government defended the tax saying it was actually a “low rated” tax since the boom would not last forever.

    This is policy and economics on the run. I hope the polls are right, let’s begin the countdown.

  11. asf

    Hmm ok, so the x-axis numbers are low because of how they define stimulus.

    Given such a small sample, I’d say the linear trendline is a little misleading. It implies that multipliers have a positive first derivative with respect to discretionary stimulus over the time period considered. How does this work?

  12. Butterfield, Bloomfield & Bishop

    asf take a read of LO.

    Gee was anyone as inaccurate as Sinkers last budget in commenting on it.

    Those estimates in the last budget were way too optimistic.

    OOPS I MADE A MISTAKE

  13. “Those estimates in the last budget were way too optimistic.”

    As are these. “Mild inflation”. It may pan out, but it is an heroic assumption.

  14. asf

    BBB I read Labor Outsider. As he says:

    ” Take out China, Korea and Australia and the relationship fades. I guess then you have to form a view on whether the Asian out-performance was related to their fiscal packages.”

    Treasury implies that it does. They say:

    “fiscal multipliers in those countries that enacted large and timely fiscal packages appear to have been larger than expected”

    So either multipliers are constant and by chance the multiplier was underestimated in countries with larger stimulus packages (in which case this is a plot of random noise) or it is implying that the multipliers increase with the size of the stimulus package (especially if you hit the magical 3% mark) and the IMF underestimated the extent of the increase.

  15. Butterfield, Bloomfield & Bishop

    No Asf his earlier stuff.
    you need to understand what is in fact stimulus and what is simply due to automatic stabilisers and a weak economy.
    Oliver Blanchard has a very good take on this in the latest IMF musings.

  16. Oh really and what does Blanchard say? Where are these comments?

  17. JC

    As Homer goes about destroying another reputation by attributing comments he didn’t make.

  18. Butterfield, Bloomfield & Bishop

    go and read them lazybones.

    It is called research

  19. dover_beach

    Homer is to research what salt is to arable land.

  20. “go and read them lazybones.

    It is called research”

    Read WHAT you insufferable imbecile?

  21. Homer,

    Blanchard’s latest is insufferable crap. Greenspan gave up targeting inflation in 2002.

  22. C.L.

    Did LO give a Boyer lecture too?

  23. JC

    Yes, he was on just before “Macker”. Homer has the transcript at home.

  24. “It is called research”

    No Homer, what you do isn’t research. It is finding a source, any source to support your position. You also plagarise and misinterpret other people’s works.

  25. In Australia’s case, causality runs from incorrect forecasts to excessive stimulus.

  26. Sinclair Davidson

    Stephen – exactly. This graph confuses corrleation with causation and the story can be reversed.

  27. Butterfield, Bloomfield & Bishop

    No Marky not that what he wrote about the WEstern world , what was stimulus, what effect the economy had , what was automatic stabilizers etc.

    for a bloke who has no clue you really should read it.

    Poor sinkers and Stephen they get it dreadfully wrong and start crying about it.

  28. “for a bloke who has no clue you really should read it.”

    I began reading the abstract and noticed an egregious error in that it characterised Western (read: US) monetary policy as rule based and targeting inflation. Greenspan gave up on this in 2002 and this is a partial factor in beginning this whole morass. The article is flawed conceptually at the base.

    As you’re not concerned with reality, you didn’t see this as a problem. Again, you find arguments from authority more appealing than facts or logic.

  29. Butterfield, Bloomfield & Bishop

    hint figure 1.7

  30. Sinclair Davidson

    Treasury are using data for the G20 for the size of the stimulus and for 2009 only. On page 36 here.

  31. drscroogemcduck

    it is suspicious that they use the ‘crisis related discretionary measures’ and not also look at the ‘overall fiscal balance figure’. lets say you pre-commit to massive stimulus before the recession then spend less then this would mean you have a negative value for stimulus in that graph.

  32. drscroogemcduck

    if someone has the time all the imf predictions for 2009 are here: http://www.imf.org/external/pubs/ft/weo/2009/01/pdf/text.pdf

    there is easy 50+ countries so having only 11 in the set means some cherry picking could have been performed.

  33. drscroogemcduck

    ah… there is only 19 countries in the document that shows the size of the stimulus. still 11/19 is kinda funny.

  34. drscroogemcduck

    if you use all 19 the coefficient is still positive much lower.

    here is my data. be aware there may be transcription errors because i had to manually copy values out of the pdf (i couldn’t even copy&paste) and i manually collated the values.

    http://www.imf.org/external/pubs/ft/weo/2010/01/pdf/text.pdf
    http://www.imf.org/external/pubs/ft/weo/2009/01/pdf/text.pdf
    http:/www.imf.org/external/pubs/ft/spn/2009/spn0925.pdf
    Balance Stimulus Predicted Actual Diff
    Argentina -1.8 -1.5 -1.5 0.9 2.4
    Australia -5.8 -2.9 -1.4 1.3 2.7
    Brazil -1 -0.6 -1.3 -0.2 1.1
    Canada -6.5 -1.9 -2.5 -2.6 -0.1
    China -4.8 -3.1 6.5 8.7 2.2
    France -5.6 -0.7 -3 -2.2 0.8
    Germany -3.7 -1.6 -5.6 -5 0.6
    India -6 -0.6 4.5 5.7 1.2
    Indonesia -1.4 -1.4 2.5 4.5 2
    Italy -4.1 -0.2 -4.4 -5 -0.6
    Japan -7.4 -2.4 -6.2 -5.2 1
    Korea -6.2 -3.6 -4 0.2 4.2
    Mexico -3.5 -1.5 -3.7 -6.5 -2.8
    Russia -13.4 -4.1 -6 -7.9 -1.9
    Saudi Arabia -10.8 -3.3 -0.9 0.1 1
    South Africa -5.6 -3 -0.3 -1.8 -1.5
    Turkey -4.9 -1.2 -5.1 -4.7 0.4
    United Kindgom -8.9 -1.6 -4.1 -4.9 -0.8
    united states -6.4 -2 -3.8 -2.4 1.4

  35. Drscroogemcduck

    That should be still positive but much lower. Like more than 80% lower.

  36. Sinclair Davidson

    your last figure for the US should be 0.311, not 1.4.

  37. Butterfield, Bloomfield & Bishop

    Have a good think of why they were not all included.

    We know overall the larger the stimulus the greater effect on the economy.

    We can also project a stimulus here of just over2% would have gained us economic growth of 0!

  38. Sinclair Davidson

    Have a good think of why they were not all included.

    data snooping? 🙂

  39. “We can also project a stimulus here of just over2% would have gained us economic growth of 0!”

    Cut taxes.

    Anyway Homer, you’re more set on ignoring facts and cherry pikcing.

  40. Well that’s lovely, Homer is “correcting” my knowledge of US monetary policy and pricing that until the GFC, the US practiced responsible rule driven monetary policy by dropping hints.

    Here’s a hint Homer – try a dose of reality. Greenspan jumped the shark in 2002. Any paper that says that austerity was practiced by the US until 2007-08 as some kind of justification to “examine” the efficacy of such “austerity” is not even worthy of you reading it, misinterpreting it, and being used as kitty litter for your cat.

  41. Skuter

    http://blogs.crikey.com.au/pollytics/2010/05/12/the-most-important-chart-in-the-budget/

    When I saw this regression (based on BP1: Box 4), something leapt out at me. The forecast error can essentially be interpreted as unanticipated GDP growth. The stimulus size is taken to be unexpected at the time of the initial forecast. Thus Sinc, isn’t performing this regression just estimating the national accounting identity Y=C+I+G+NX? It was the fact that the coefficient was so close to 1 and significant that thought of this as a possibility. Your (and anyone else’s) thoughts?

  42. Skuter

    Whoops

    “…that I thought of this as…”

  43. JC

    Skuter:

    You can’t make those comparisons shown in the Crikey column because it is not taking into the account the individual shock incurred by each nation.

    It’s assuming everyone was even keel going in. they weren’t.

  44. Skuter

    JC, very true. There are many reasons to look at this ‘analysis’ and laugh. It just doesn’t pass the sniff test. What about omitted variable bias – the monetary policy actions taken by central banks did nothing (or is Treasury somehow saying they were anticipated changes whereas the fiscal changes were not)??? In my previous entry, I am just merely positing that the results are merely a statistical artefact – the result of estimating an identity…

    As for the attempt to extrapolate to a no stimulus case, I just ignored this as it is rubbish of no value to anyone.

  45. Sinclair Davidson

    Skuter – maybe. I’ve been wondering about the truncated sample.

  46. The AIC on possum’s model is mediocre and the R -squared is very low. He’s modelling a spurious relationship.

    “What this chart demonstrates is that the size of the stimulus packages mattered – seriously mattered. The larger the stimulus package, the better a country performed compared to what its original IMF growth forecasts suggested.”

    That’s all he proved.

    Here’s what really happened:

    Net arrivals -12600

    Changes in hours worked – negative, equivalent to -1.4 unemployment or eqivalent to 169 200 job losses

    Increase in discouraged workers, Sep 08-Sep-09: 37 900

    Increase to “not marginally attached to the labour force”, Sep-08 to Sep-09: 157 700

  47. JC

    Yea… The amusing thing is that monetary policy is invariably never mentioned by the stimulus peddlers as an important lever. In fact when it’s raised they either get very cranky or dismissive about monetary policy and usually start mumbling incoherent angrified swill.

    To me what doesn’t pass the smell test is to believe that the Lurch/Rudd insulation progress fiasco kept the country from going into a deep and long lasting depression. 🙂

  48. Pingback: Did the stimulus work? at Catallaxy Files

  49. Monetary policy and a floating exchange rate saved us in the face of market and policy calamities. The currency depreciated by 30% when our export prices had fallen.

    If we had a fixed exchnage rate and more inflexible labour market, unemployment could have been disasterous.

    Supporters of the fiscal stimulus and Keynes should realise two things: 1. we weren’t in a liquidity trap so the stimulus wasn’t justifiable, 2. The monetary expansion was equivalent to about 30 bn of stimulus. Thankfully were winding it back.

  50. Sinclair Davidson

    Refresh and read the new post.

  51. JC

    It’s the steepest exchange rate move I’ve ever seen. Even compared to the time of the float we’ve never moved so quickly.

    To its credit the RBA essentially left it all alone and let the exchange rate clear and reflect the impact of the adversities hitting us at the time.

  52. asf

    Possum should give up on the statistical analysis. Whoever gave him that EViews licence has a lot to answer for.

    Mighty brave of him to publish the regression output and then claim the results are meaningful.

  53. Sinclair Davidson

    Possum’s a good guy.

  54. Butterfield, Bloomfield & Bishop

    Marky the hint is about how much of Sinkers ‘stimulus isn’t stimulus at all about the economy affecting the budget.

    If monetary policy saved us the the interest sensitive sectors would have risen togetheralah 1992
    They didn’t the main sectors that rose were those affected by the stimulus.

    no we don’t have a liquidity trap like most other Western countries but our monetary policy was impaired by the credit crunch.

    Our banks lacked capital.

  55. “no we don’t have a liquidity trap like most other Western countries but our monetary policy was impaired by the credit crunch”

    ???

    Do you understand how high powered money works? It worked perfectly well.

    “Our banks lacked capital.”

    Sure, that’s why private construction didn’t fall in a heap. That’s why they didn’t go under.

    “If monetary policy saved us the the interest sensitive sectors would have risen togetheralah 1992
    They didn’t the main sectors that rose were those affected by the stimulus.”

    The construction sector is interest rate sensitive, you dummy.

  56. Butterfield, Bloomfield & Bishop

    oh dear which parts of the construction industry.Why for example didn’t most parts of the housing industry rise together as in 1992.
    you really are quite ignorant aren’t you.

    Monetary policy affects then all not just first home buyers.

    Our banks even with a government guarantee could access overseas markets until the worst of the GFC was over.

  57. “Our banks even with a government guarantee could access overseas markets until the worst of the GFC was over.”

    You’re point being? If you were literate, you’d have a point by now.

    “Why for example didn’t most parts of the housing industry rise together as in 1992.
    you really are quite ignorant aren’t you.”

    No I’m not shit for brains. All property is facing high demand and low supply, resulting in price growth except for warehousing.

    Gutless spiv.

  58. Pingback: Did the stimulus work II? at Catallaxy Files

  59. Pingback: Heritage Foundation on fiscal stimulus at Catallaxy Files

Comments are closed.