Recently, new national accounts guidelines have been published. The national accounts distinguish, within states, several sectors: the Government, Non-financial companies, Financial companies, Non-profit institutes Serving Households (from churches and commons to your local amateur soccer club) and Households. Today: What do the new 2025 guidelines state about Households? Do they advise estimating inequality? Do they pay attention to households as producers of market and non-market goods and services? Are there important conceptual issues to consider? These and some related issues will be discussed below.
Households, generally being the largest spenders in the economy, have been at the centre of national accounting for a long time. From the outset, one of the key issues was the distribution of income among households. In this series, the work of Mitchell e.a. in the 1920s about this has already been mentioned. On top of this, in ´How to pay for the war´ (1940) Keynes introduced, for analytical reasons as well as because of domestic fairness, a novel, estimated concept of the national accounts (pp. 81-82). His main innovation: a division between rich and poor UK households, including data on the primary sources of income and expenditure. More recently (2023), Berman and Milanovic used the concept of ´homoploutia´ or ´the situation in which the same people are rich in the space of capital and labour income´. One of the sources they use to do this are the micro files of the distributional national accounts of the USA, as published by Piketty e.a. (I did not check these data, the link on the internet was broken).
But for a long time, the road taken by Keynes and now trodden by people like Berman and Milanowic as well as Piketty e.a. was not followed by national accounts statisticians. Distributional accounts consistent with the national accounts are rare. Piketty, Saez and Zucman (2016) explicitly try to fill part of this gap for the US, thereby making the remaining potholes visible. And the new guidelines state in the introduction of the chapter about households (lemma 32.5):
´There are many different types of households that can be formed, including those grouped by income decile, those undertaking unincorporated production activities and those whose members have specific characteristics (e.g., in terms of number or age of individuals).´
Following this, the guidelines strongly endorse distributional accounts. Heterogeneity, including differences in income and wealth, is viewed as a fundamental aspect of the household sector as a whole. According to the manual, there are non-linear economies of scale related to the size of households. Additionally, changes in the level and distribution of income not only lead to changes in the level of macro-expenditure but also in the composition of macro-expenditure. Taken together, this necessitates a ‘sub-sectoring´ of households, which not only distinguishes different types of households but also requires distributional accounts. However, the manual also points out some fundamental problems with this approach. For one thing,
´most income is earned by individuals but Consumption is undertaken by households´.
Additionally, according to the guidelines (drafted by experienced statisticians), it is challenging to combine different categorizations into analytically homogeneous sub-sectors. Making a distinction between capitalists (owners of assets) on one side and ´labour´ (people receiving a wage) on the other runs into trouble when people with high labour income also have high capital income, as mentioned by Berman and Milosevic. Despite such difficulties, the renewed insistence on distributional accounts is welcome and tallies with the work of economists like Mitchell e.a., Keynes, Kuznets, Piketty et al., as well as Berman and Milosevic and many others.
Another point is the concept of Consumption. The national accounts define, except in the case of houses, the monetary value of Consumption as the purchase of goods and services. It is clear that in the case of households, many purchased goods serve as intermediate goods used to produce the ultimate consumption goods and services. This also needs skilled labour. Take the 24 steps required to make a tarte-tatin, as I will soon have to do. Running a household requires skills and knowledge about hundreds, if not thousands, of such procedures. This means that consumer durables are a type of fixed capital with embedded technology, while also bringing unpaid household labour into the focus of discussion.
In my view, the national accounts, however, must track the money economy, meaning that household monetary purchases are a meaningful concept. This can be complemented with household balance sheets, including separate accounts for the amount, quality, and value of houses, cars, and home appliances. Prices on second-hand markets for houses, cars, etc., are readily available; there is no need at all to use, as the guidelines suggest, discounted values. However, ´raising the kids,` even when more critical than most or even all of the activities tracked by the national accounts, might be less suitable as an item for the national accounts. Even when many of the services and goods needed to do this are tracked by national accounts and related statistics, like the inflation data. The definition of these goods and services might be adapted to enable use in auxiliary household accounts.
Households can also be producers in the sense of the national accounts. Traditionally, this is often the case, as the guidelines state, in agriculture. Nowadays, possibly somewhat neglected by the guidelines, the production of electricity by households is non-trivial, while over the last few decades, Bed & Breakfast and comparable services have turned out to be a growing and often profitable activity. In combination with the technological progress embedded in household appliances (such as cars, washing machines, mobile phones, and LED light bulbs), a technology index for household machinery could be created. In production sectors, labour productivity (total value added divided by hours) is often used as an indicator of technological progress. In non-monetary or only slightly monetary sectors, such indicators should not be used. Cars owned by households are idle for 95% of the time. Meaning that these are used in a totally different way than cars owned by, for instance, a taxi company. Companies like Uber are changing this. But just a little. Household production has a different logic from market and monetary output.
These subjects are far from new. But it is good to see that they gain renewed importance in the new guidelines.
