In summary, using household budget surveys with a large number of respondents show that while male wage rates underestimate total household income, they seem to work pretty well for cross-country comparisons as long as we use urban wage rates. To read more you can follow this link shorturl.at/VvtHu

Using calorific availability as a proxy for living standards, I show that Swedish urban and rural households were very similar, and they were also very similar to British urban households. This is also what we find in traditional real wage studies comparing Sweden and Britain in this period.

This means that if we use urban and rural male wages, the income shares differ between countries, but not if we only use urban wages. But how representative are urban wages? What about the urban-rural wage gap? It turns out that it was very small when accounting for total household income.

What about urban-rural differences? Scandinavia was much more rural than Britain at this time. Unfortunately, only Sweden and Denmark included rural workers, but for them, the wage of the household head constituted a much smaller share of total household incomes, roughly 70%.

Instead, I used early 20th century household budget surveys from Scandinavia and Britain to see if relative standards change if we include incomes from other sources than the wages of the male household head. As this figure for urban households show, additional incomes were similar across countries.

This can certainly be an issue, but if the bias is similar across countries, real wages can still be used for cross-country comparisons. The only recent study that has looked into this indicates that countries differed in additional income shares. This is based on quite few data points however.

The background is that several papers have argued that male real wages are not reliable as indicators of living standards as they are often based on day rates from one single sector and doesn't include additional incomes from other family members.