One of the clips of Thatcher which has been passed around in the days since her death is this one, of two exchanges from her last speech in the House of Commons on 22 November 1990. In it, she attacks the left-wing focus on equality by arguing that that focus ends up resulting in making the poor poorer – just by less than it makes the rich poorer:
I've never understood why it's an exchange held in quite such high regard – the bit at the end, where she starts making graphs with her hands, is a particularly excruciating bit of political communication – but the point does stick home. It's not that common to hear arguments for the poor to be made poorer, but it remains the case that policies which help both rich and poor are argued against on the grounds that they help the rich more.
There's good reasons for this, of course. As Richard Wilkinson and Kate Pickett's book The Spirit Level documents exhaustively, in developed nations like our own, a huge number of social, political and health outcomes are dictated by equality, not absolute wealth. So even given the fact that Thatcher's legacy was of the poor getting richer, the fact that that it included a massive increase in inequality  may have meant that poor people were worse off at the end of her premiership than the beginning.
But if hurting the poor to hurt the rich more is a trap for socialists, there's a sort of parallel problem that Conservatives fall prey to: an opposition to redistribution which prevents them enacting policy combinations that help everyone.
A new paper by economists David Autor, David Dorn, and Gordon Hanson, titled "Untangling Trade and Technology: Evidence from Local Labor Markets ", compares and contrasts the effects of trade and technology on employment. On the face of it, it doesn't matter to you whether you lose your job because a robot can do it cheaper, or because a Chinese labourer can do it cheaper: you still don't have your job, and your employer has more money. But in actual fact, the two have markedly different macroeconomic effects:
Trade exposure reduces overall employment and shifts the distribution of employment between sectors, [but] exposure to technological change has substantially different impacts, characterized by neutral effects on overall employment and substantial shifts in occupational composition within sectors.
Trade in particular is found to impose "particularly large" employment losses on workers without college education; but even technological change, which is neutral on "overall employment", has the effect of destroying middle income jobs while bolstering high- and low-paid labour.
Of course, all of that is background to the strong evidence that technological change and free trade make society as a whole richer. The problem isn't with the lack of gains – it's with the distribution of those gains.
How do you deal with good gains and a bad distribution? You bank the gains, and fiddle with the distribution. Cut taxes – or boost tax credits – at the bottom end of the income distribution, and pay for it with higher taxes at the top end. Or even just leave taxes at the top the same, and use the fact that the rich are getting richer – and thus paying more tax – to (more than) compensate the poor for the losses.
This is the lesson that Thatcher, and the Conservatives who have followed her, never learned. It's more than just economic good sense: it's politically useful, too, to be able to tell everyone that they will be made better off. Think how much easier the debate over immigration would be if the Tories could point to a tax cut – for the poor – which was funded through the increased gains migration brings.
In the language of economics, there are very few pareto-optimal policies left; the number of changes you can do which help everyone, as opposed to helping some and harming others, has dropped close to zero. But a good bundle of policies can still make everyone better off – and that bundle will nearly always include redistribution of wealth.