Keynesian Multiplier of 0.05

So much for that Keynesian stimulus notion (emphasis in the original)

With everyone focused on the 5th anniversary of the Lehman failure, we are taking a quick look at how the world's developed (G7) nations have fared since 2008, and just what the cost to restore "stability" has been. In a nutshell: the G7 have added around $18tn of consolidated debt to a record $140 trillion, relative to only $1tn of nominal GDP activity and nearly $5tn of G7 central bank balance sheet expansion (Fed+BoJ+BoE+ECB). In other words, over the past five years in the developed world, it took $18 dollars of debt (of which 28% was provided by central banks) to generate $1 of growth. For all talk of "deleveraging" G7 consolidated debt has been at a record high 440% for the past four years.

The theory of stimulus -- taking money out of the productive economy, where it is spent based on the information of hundreds of millions of people as to the relative value of millions of potential investments, and handing it to the government to spend based on political calculus -- never made a lick of sense to me.  I guess I would have assumed the multiplier in the short term was fractional but at least close to one, indicating in the short run that if we borrow and dump the money into the economy we would get some short-term growth, only to have to pay the piper later.  But we are not even seeing this.


  1. ErikTheRed:

    Because Keynesian spending is virtually always malinvested, if you consider opportunity costs the multiplier is deeply negative.

  2. Ben:

    While I agree with the conclusion, the baseline of "exactly 0% GDP growth" needs some defense. Stimulus advocates would surely say that GDP growth would have been (very) negative but for the stimulus, and therefore the stimulus was successful. It's the same reasoning about how the stimulus was awesome at creating jobs even though not many jobs were created, because the job loss but for the stimulus would have been huge. It's what makes the impact of stimulus basically unfalsifiable.

  3. Max:

    I actually thought the total opposite. I have seen public works projects and even in the relatively efficient countries like Germany, they take a long time to take off the ground (and never on budget). So in all earnestness, you can already take out government projects from the Keynsian spending. So, the trickle down money and work that should jump-start the economy will most likely arrive when the economy is already back in business (5-6 years out at least). This leaves the portion that is tax credits and credit lending. Tax credits might help, but if the underlying regulation scheme is not changed they only mean a shifting of purchases from the second year to the first year (like cash for clunkers). Then we have credit lending and here we see only effects in Wall Street and not main street, because businesses like your business were vary of its effects. Most serious businesses didn't believe in the magic power and they were perfectly right not to. This lead to a jobless recovery (also ACA didn't help - even a single payer system might have been better).

    So, actually, I thought that maybe 5 to 6 years later there would be some bump, but even that seems not to happen now. Although one could make a point that Germany had a positive Keynsian stimulus coefficient. However, I think that has more to do with the unique situation in Germany. The Green and Social Democrat union in the early 2000s actually was pressured to relax labour laws and thus allowed for more flexible actions later on. They allowed companies to reduce pay and working hours during the crisis situation, allowing to retain most of their workers to start over once the crisis ended in Germany. When I read about it, I thought isn't that a nice way to deal with sticky nominal wages? Of course, something similar was impossible in France, where a deeply entrenched syndicate network prohibited any kinds of changes in the labour market. The result today is that the big production centers of Renault and Peugeot in France are in big trouble. Peugeot shed 1000s of workers and will probably have to close one plant. Renault is currently supported by its big financial subsidaries and its Japanese ally Nissan. However, it shows that those two french behemoths are two of three companies that lose money on each car sold (Fiat is the third). So they are not really successful in their core business. In contrast the Nissan gains 800 € per car sold and Toyota ~2000 €/carsold. And those two are mass market competition, not like Mercedes or BMW luxury brands.

    So here you see labour markets and its effects in greater detail at works. You also see the effect of over-regulation and the promise of change. Though I think I got a bit off-topic here.

  4. bilejones:

    "positive Keynsian stimulus coefficient. There is no such thing.
    It merely moves wealth around
    The creation of the money needed for the "stimulus" for the past 100 years has always benefited the banks, the first to receive the new money and penalised the poor, the last to receive it.

    Keynsian stimulus always and everywhere concentrate wealth in the hands of the financial class.