Zarah Sultana's recent claim that we have suffered 40 years of Thatcherism prompted Blairites to point to a big difference between Thatcher and New Labour - that whilst inequality rose a lot under Thatcher it largely levelled off under New Labour.

Measured by the Gini coefficient, this defence is correct. According to IFS data for incomes after housing costs, this coefficient rose from 0.261 to 0.374 from 1979 to 1996-97, but edged up only slightly more (to 0.405) during the New Labour years.

What should we make of this? A bit of me recalls Chris
Rock: "you want some credit for some shit you're supposed to do."

Also, though, the Gini coefficient is, essentially, an average of all inequalities. It can stay the same if inequality increases in some respects but falls in others. And this is what happened under New Labour.

My chart, taken from IFS data, shows the growth in real incomes after housing costs, for each fifth percentile of the income distribution. It shows that it was the 20th-30^th percentiles that did best under new Labour - that is, those worse off than three-quarters of people. These are those with incomes after tax and housing costs equivalent to around £280 a week today for a childless couple.

Inequality tended to fall under New Labour because these got better off relative those on slightly higher than median incomes. This wasn't just because of market forces. New Labour did genuinely help the moderately badly off with policies such as the minimum wage and tax credits.

Other inequalities, however, increased. One was between the very lowest incomes and this 30^th percentile. Some of these shouldn't be the object of our sympathy: they are wealthy folk who are living off their capital. Others, though, are those who have fallen through the safety net or are unemployed.

The other inequality that increased was at the top end. The 95^th percentile - equivalent to a weekly income of £1160 in today's money - got even better off than the median. These IFS data don't cover the very rich. But HMRC data do. And they show that the top 1% of taxpayers also did well under New Labour, with after tax incomes rising 56% compared to 38% for the median taxpayer. Peter Mandelson was famously "intensely relaxed about people becoming filthy rich". And they did just that.

The flattish Gini, therefore, disguises the fact that some inequalities fell whilst others rose.

Which poses two questions.

One is: how does the Tory government compare to New Labour?

My second chart helps answer this. It shows a similar pattern in one regard: those slightly below median incomes have done well relative to both the very worst off and those on above-median incomes. But it also shows that the very best off have seen their incomes fall. Egalitarians should welcome the latter at least.

Perhaps the biggest difference with New Labour, though, is that the Tories have presided over little growth in incomes across the board. The group to have done best under the Tories are those around the 35^th percentile (on £340 a week in today's money). Their incomes rose 5.3 per cent in real terms between 2009-10 and 2017-18. But this percentile saw income growth of 38 per cent under New Labour.

The second question is: which inequalities matter most? It is here, I suspect that New Labour was at fault.

The equality that it increased was that between those a bit below and those a bit above the median. This, though, is a second-order problem. People on moderately poor incomes can live almost as well and those on moderately decent ones, at least with a bit of careful budgeting. Sure, they'll have smaller houses and worse cars. But their spending on everyday items, and on nights out and even holidays, isn't very much different.

Instead, what matters more are inequalities at both extremes. Very low incomes mean genuine distress; having to forego essentials; being only one accident away from catastrophe; the uncertainty about wondering how many hours you'll get to work next week; and the cognition-sapping stress (pdf) of just getting through the week.

And at the other end, it is the increase in very high incomes (and the power that lies behind them) that has contributed to the slump in productivity growth. What's more, such inequality undermines democracy as the ultra-rich leverage their wealth - and the undue deference they get - to acquire disproportionate political influence.

So, yes, Blairites are right: New Labour did greatly slow down the increase in equality we saw under Thatcher. But - in hindsight - it did not do enough to tackle the inequalities that most matter.