That is interesting. I confirmed those numbers looking at the actual census data. The changes you mention are much less pronounced if you remove no-income households from the data though. Perhaps no-income households should be included in the 2 or more stat - since one could make a case for that category being the people who "gave up" because they couldn't keep up with expenses. I wonder what other things might be unaccounted for that could make the data "seem" this way - or perhaps my anecdata is just wrong. Also, the data doesn't go back further than 1980. I wonder what it would show if we had it back to 1970.

[Edit] Just another thought - that data does not break down the households into percentiles for total household income. I would expect that the fraction of multi and no income households w.r.t. total households has increased more on the lower income side of the spectrum than the middle/upper end. It also doesn't account for how much each earner brings in - which is related to the weekly hours stat in your source - I'd expect this to be even more influenced by accounting for income quintiles (or quartiles).

I wonder if those single income households include maintenance and alimony. The nature of families has changed significantly since the 1970s with more divorce and more children being raised in one parent households. I doubt the single parent receiving money from the estranged parent is counted as a two earner household, even though in some senses it is.

Once you add the changing rate of marriage, increase in single person households since the 70s, and the fact that real inflation often varies depending on which income quintile you're in, I don't think enough data is available to let you pick it apart meaningfully. So we're left trading anecdotes or stats with gaps. :)

The number of traditionally shaped families, i.e. mum, dad, and 2.2 children has plummeted since the 70s and the number of singles, with or without children has rocketed from both increasing separation and later entry into relationships.

I'm in agreement with your main point, that it's far harder for a single earner family to thrive now than in the past. I doubt I can cite that in any sensible way.

But has it changed in recent years? Anecdotally, I was born in 1983 and when I was growing up in the 90s single-parents & working-mothers were already common.

>the number of 2 or more income households has actually fallen

Where in your citation is this claim supported? The /percentage/ has fallen which could either mean that the number has fallen or that the total population increase at a faster pace than the number.