We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Samizdata quote of the day – peddling long disproven nonsense

Just to make the problem clearer. Economies do add up. If this happens here then that over there must also happen. If we don’t see that second then we’re mistaken in our assumption that the first has. If productivity has risen and wages haven’t then the labour share must have fallen. The labour share – up to when PK wrote in 1996 – had not fallen. Therefore that confident blue line from 1970 to 1996 is wrong.

We don’t even have to worry about why it’s wrong. It just is – so bollocks to the rest of it.

Chakrabortty’s getting on a bit to be an enfant terrible of course, his unwillingness to spend time and energy understanding the economics he’s attempting to write about is easier to explain for he’s at The Guardian. In fact, he writes the economic editorials for The Guardian and an actual knowledge of economics in that job – let alone time and effort spent gaining it – would be a positive hindrance.

No, really.

Tim Worstall putting the boot in 😀

9 comments to Samizdata quote of the day – peddling long disproven nonsense

  • Johnathan Pearce

    So let me get this straight: the blue line in the graph that the Guardian writer uses is plain inaccurate and based on something he pulled from his arse?

    I am not sure that Tim Worstall’s delightfully brutal response is going to work on the sort of people who read this article and think, blow me, this shows how terrible Western capitalism is and how it screws the workers. What I’d like to see Tim or someone do is put up a graph that does, actually, capture what the trend of real wages has been.

  • Paul Marks

    There are two approaches to economics – the logical (logical reasoning) approach of such economists as Ludwig Von Mises, and the empirical approach of such economists as Milton Friedman.

    Done correctly both approaches reach the same place – that state intervention, like the intervention of private criminals (such as the gangs who can take everything you have in some Latin American countries), does harm – not good.

    Guardian writers, and the international establishment generally, pretend to use the empirical approach – but they do not really do so – as they rig their “studies”.

    Yes Johnathan Peace – they LIE.

    The most “respected” government and academic bodies lie – and lie without shame. And why not – after all American “Pragmatist” philosophers “proved” more than a century ago that either objective truth either did not exist or humans could not reach it – so all that matters is POWER, the Progressive agenda.

    And this does not just cover economics – they lie about the natural sciences as well. They lie whenever they “have to” – i.e. whenever lying helps their agenda of power and control.

    Plato’s Guardians were to do the same – and he considered their lying to be noble.

  • Sigivald

    Did not expect Paul Krugman to appear, nor to be speaking sense.

    But then, he DOES actually know his econ, when he doesn’t need to bend it to politics – and 30 years ago was back when he was an economist, not a pundit who was previously an economist.

  • Paul Marks

    Sigivald – Paul Krugman lies, he is not making innocent mistakes – he lies.

    They all do.

    These “empirical economists” will tell you, with a straight face, that (for example) under President Herbert Hoover free market policies were followed – and failed.

    In reality there was a massive increase in tax rates and, for the first in American peacetime history, a government campaign to PREVENT wages falling to adjust to a Credit Money bust.

    They will tell you that laissez faire policies were followed in Ireland in the 1840s – not mentioning the crushing Poor Law taxation.

    They lie about “the data”, the historical fact, in every case – they LIE.

    Ludwig Von Mises argued that the empirical method is not a good way of doing economics – but these establishment types are not even using an empirical method.

    Basic rule of using an empirical method – do not lie about the “data” (the historical facts) – but that is what they do.

  • Jacob

    The claim “they lie” can be made only if measured against some objective standard that all agree is “true”.
    But there is no such thing in economics. It’s all doctrines and ideology, and, as Mencius said: – on every issue you’ll find economists on both sides of it.

    As for the issue at hand: the question is if you believe statistics about “labor share”. For example – if you count all government and public employees (such as, say, teachers) — then yes – “labor share” can grow (by striking and getting pay rises), and GDP too (their salaries are counted), without a “positive” or productive economy. I mean: economic statistics are as fluid as economic theory.
    It is complicated. You cannot claim, like Tim – “if A then necessarily B”.

  • Johnathan Pearce (London)

    So, to repeat my question, what should the graph that this Guardian writer used look like if the data used is not distorted?

    It is always good to confound bad visual representations of what is going on with better ones.

    Tyler Cowen had these observations back in May this year:

    There is some bad news afoot for workers. Labor’s share of the US gross domestic product has been falling for a long time, by seven percentage points since World War II. The labor share for 2022 — depending on exactly which measure is used, it comes in at slightly more than 60% — is the lowest measured since 1929.

    And it’s not just America. Globally, the labor share, which is the fraction of an economy’s output that goes to workers, has declined by six percentage points since 1980. The numbers suggest that the share of labor is declining in 13 of the 16 wealthiest countries in the world.

    One possible explanation for labor’s declining share is simply that the cost of capital has been falling for decades in most countries. That development benefits capital income very directly: It’s cheaper to raise capital, which benefits workers only indirectly. Of course, with real interest rates higher recently, it will be possible to test whether the labor share of income will make a comeback. In any case, this stands as one of the most plausible hypotheses.

    Globalization and automation are two other trends that may have made labor markets more competitive, at least as compared to capital markets. Yet it is not obvious why those forces would lower labor returns more than capital returns. Is labor more mobile internationally than capital? Even if you think US companies have benefited from buying cheap manufactured goods from China and then reselling them at the expense of US workers, that doesn’t explain why labor’s declining share has been so widespread across countries and decades. If globalization were the culprit, labor’s share should be rising in China and other major exporting countries — but the opposite is true.

  • Fraser Orr

    @Johnathan Pearce (London) quoting Tyler Cowen:
    There is some bad news afoot for workers. Labor’s share of the US gross domestic product has been falling for a long time, by seven percentage points since World War II. The labor share for 2022 — depending on exactly which measure is used, it comes in at slightly more than 60% — is the lowest measured since 1929.

    One thing this doesn’t seem to account for — something that seems fairly obvious to me — is that perhaps labor’s share of the GDP has been falling because labor has been contributing a smaller fraction to GDP? To give a simple example: in 1929 all bookkeeping for all companies was done by hand in huge paper ledgers. Now 90% of that work is done by computer. Or on a Ford assembly line thousands of men bolted and welded together cars which is now done by robots. Which is to say small amounts of labor are amplified by large amounts of capital. So that capital investment leverages the work of workers, but if the lever, bought with capital, does most of the work, it hardly seems reasonable for the lever operator to demand the all value that that capital produced.

    Of course what isn’t said is that even if the laborer’s percentage amount decreased, his actual amount dramatically increased. Everyone from the CEO down to the janitor to the guy screwing bolts on the assembly line (in fact down to the guy sleeping rough in the streets) is vastly better off than the same guy fifty years ago. So which is more important: being richer than your neighbor or being rich enough to enjoy a happy fulfilled life?

  • bobby b

    If a society brought in tons of cheap cheap labor – labor that was willing to do the construction and mowing and programming and roofing and driving and retail and nursing and whatnot for about half the previous going rate – wouldn’t the labor share look just like that graph?

  • Tim Worstall

    There is an estimation out there of the real chart. Can’t find it this morning. But Cato, someone like that, has done it. Productivity and total wages have tracked each other. It is though true that “average production worker wages” haven’t tracked so well given the increase in wage inequality

    The capital share thing from Tyler. Well, yes. Sorta. The labour share has been falling in the US but from the 1990s to now. So Krugman was right – and why I quoted the specific years. But, and here we get into knotty problems. There are actually four components of the national income. Capital, labour shares. Then “mixed income”. Independent trades, that sort of thing. Is $5k of tools capital or not? So, we don’t try to allocate, we call it that mixed income. Finally “subsidies to production and taxes upon consumption” as the fourth. The UK reports the four separately and it’s possible to then see something interesting. The labour share has fallen, yes but the capital share hasn’t (other than recovery from the impossibly low levels of the 1970s). Mixed income has risen and so has that fourth – all that green spending is in there, so’s the doubling of the VAT rate since the ’80s and so on. In the US sales taxes have got higher and more places have them over the period.

    The labour share has fallen. But it’s to tradies and government, not capital. Which is interesting.

    The US doesn’t do it quite like that. Census (think, could be BES) allocates the last two over the first two. So changes in that portion of the economy which is mixed or tax plus subsidy gets allocated to capital and labour. Which is OK – except when it’s that very difference which we want to explore and we’ve now obscured by using the same old weights to allocate over labour and capital.

    There’s also one more wrinkle. I actually have difficulty believing this but I did run it to earth once and this really was what – I am sure but am always open to correction – they were doing. One of the differences between GDP and GNP is how to treat the foreign profits of domestic firms.

    For GDP include domestic profits of foreign and domestic firms, exclude foreign profits of domestic firms. For GNP exclude domestic profits of foreign firms and include foreign profits of domestic firms. D means what happens in our economy, N means what flows to the people of our country after all.

    But, from what I could see – and still don’t quite believe – the US GDP numbers include the foreign profits of US firms. Which is terrible economics but that’s what – again as far as I could see – they do. And that’s a problem. The overseas profits of Apple, Google. MSFT and so on do add up to a percentage point or two of the whole US economy. Skews the ratios considerably.

    Of course, there’s an easier way of putting all of this – economic statistics aren’t very good. Therefore we can’t use them for detailed proofs or management. But then Hayek got there before us on that point, no?

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>