A little taxonomy of inflation – there is no thing like ‘the’ price level

Technical addendum: ‘Individual consumption expenditure of general government’ equals spending on health care and education and the like. Collective expenditure equals the proverbial streetlamps. NPISH stands for Non Profit Institutions Serving Households like churches, unions and soccer clubs.

At this moment there is quite some talk about the specter of inflation. And indeed: some prices are increasing. Houses! The runaway increase of house prices sure is a specter to be bothered about! But should we also bother about ‘flow’ prices, like consumer prices or the prices of fixed investments? The answer is: life isn’t simple. There are many kind of ‘flow’ prices. We should look at them in tandem. The consumer price index has been developed and designed to estimate the purchasing power of households, meaning that increases of the consumer price level should be analyzed together with increases in wage levels, other income components and hours worked. Focusing on one price index only, as central banks tended to do (they look at many, but policy is focused on one), is not very helpful. Surely not during crises when large sectoral differences lead to large differences between different price indexes (figure 1). The lock downs have, directly or indirectly, increased costs, led to disruptions of supply chains and a surge in hiring leads to shortages of labor in some places. Don’t bother. Or: do bother, but that means that we should not bother about erratic price developments but about the ability of households and companies to survive the disruptions.

The erratic price developments have to play out – also to enable companies to survive, to look at it from an Austrian angle (we have to curb house prices, however). Which does not mean that we do not have to look at these prices. We have to look at them. But figure 1 shows that there, indeed, are quite some different developments when it comes to different categories of spending. We do not really understand the consequences, at the moment. But we have to say goodbye to focusing on one price index only. Inflation is understood as a rise in the general price level. Unfortunately, even when we do estimate something like a general price level – the GDP deflator comes close – economist often look at the consumer price level as ‘THE’ price level. Which is wrong. There are more kinds of spending than consumer purchases alone. Each of these has its own price level. And it’s not even about expenditure alone. The national accounts distinguish production accounts, spending accounts and income accounts. Each of these has its own price levels. A taxonomy looks a bit like this (leaving financial expenditure, like emitting stocks or purchasing stocks aside):

Expenditure price levels

• ‘flow’ expenditures
  • Household consumption expenditure
  • Public consumption expenditure
    • Individually consumed (aka education)
    • Collective consumption (aka the proverbial streetlamps)
  • Investments in fixed assets
    • New houses and other buildings (excluding land)
    • Public infrastructural investments (roads, bridges, excluding land)
    • New military equipment
    • Private investments in fixed assets, including Research and Development
  • Exports
  • Imports
• stock expenditures
  • Purchases of existing houses (including land)
  • Purchases of second hand machinery, military equipment, cars and the like

Income price levels

• Wages
• Rents (land, houses, leasing cars or military equipment or even soccer players)
• A special category is ‘profit’: being a purely financial category this has no ‘unit’ in the sense that wages have a natural unit (hours) and hence has no price level

Production price levels

• Industrial producer price levels (there are many of these)
• Prices of farm products
• Prices of services, including banking fees
• …

You get it: there are more prices between heaven and earth, Horatio, than are dreamt of in the philosophy of the representative consumer models. These, distinguish only one product and as there is not really any kind of division of labor in the one person representative consumer models also do not make clear distinctions between production, consumption and income. Truth has to be told: more modern HANK models (Heterogenous Agent Neo Keynesian models) do enable a distinction between, for instance, laborers and capitalists and hence a distinction of wages and rents and house prices and the like (even when HANK models too are bad when it comes to operationalizing all kinds of variables). Inflation targeting by central banks, which still has not been formally abandoned, is however still based upon one price level and implicitly on the representative consumer model. Don’t go for it. Life is complicated.