Ed Miliband's move to 'mend not end' Labour's relationship is a brave decision - both in the conventional sense of that term and in the sense famously employed by Yes Minister's Sir Humphrey Appleby, namely to suggest that a particular course of action is, shall we say, 'not entirely free of risk'.
That presumably is why, as he admitted soon after Mr Miliband's landmark speech earlier this week, even Tony Blair, who, after all, made something of a habit of taking on the 'forces of conservatism' on the left, as well as on the right of British politics, decided that, on this issue at least, discretion was the better part of valour.
The risks of distancing the Party from the unions are clearly not simply financial. The labour movement arguably anchors Labour to the interests - or at least the perceived and the collective interests of ordinary working people of the country it one day hopes to govern again. More directly, trade unions, by providing volunteers and other 'in-kind benefits' to Labour during elections, form an integral part of the Party's campaign infrastructure.
Indeed, it's probably no exaggeration to say that without union help in 2010 Labour would not have been able to hold on to anything like the number of seats it eventually did. Moreover, whatever emphasis the Party puts on 'community organising' in the run-up to 2015, its hopes of replicating Obama's successful voter-identification/mobilisation operation at the next election depend in no small part on trade unionists willing to do the donkey-work without which even the cleverest digital strategy won't make any difference.
Still, the money matters. Which is presumably why it's the money that the leaders of the big unions have been dropping dark hints about in the last couple of days. According to the BBC, Len McClusky told his Executive yesterday that Unite, whose one million affiliated Labour members currently contribute around three-and-a-quarter million quid direct to the Party 'believes the number of members affiliated to Labour could drop as low as around 75,000 after the changes - reducing fees to around £250,000.' Earlier in the week, the GMB's Paul Kenny told the Beeb that its affiliation fees could drop to a pitiful ten per cent of today's total if members are required to opt-in to paying the political levy that goes to Labour rather than opting-out.
Estimates as to how much Labour could lose from Mr Miliband's plan, presuming it goes ahead in the way that most informed observers currently think it will, vary considerably but the most oft-quoted figure is a cool (millions are always cool) five million pounds.
Those educated guesses - and that's all they are, of course - are ultimately based on estimates of how many union members would consciously choose to join Labour as affiliated members. Quite what people are basing those guesses on, I'm not sure. Indeed, I'm not even sure how we should be doing it.
Certainly one way to do it - in fact, one sure-way of assuming the worst - would be to assume that the proportion of trade unionists willing to opt in will broadly reflect the proportion of people in the population as a whole who are members of political parties. Since in the UK that is about one (yes, you read that right, one) per cent, there will be many on the centre-left who worry that Mr Miliband is heading for a huge cash-flow problem and is therefore cutting off his nose to spite his face - or at least to spike the guns of his Tory opponents in parliament and in the media.
But not so fast. For one thing, trade unionists do still disproportionately incline, firstly, towards joining things and (to a much lesser extent) towards Labour, too. So it would be reasonable to assume that things won't be quite as bad as all that. For another, there is an alternative measure we can refer to. It's a historical one, but perhaps none the worse for all that: to paraphrase an iconic Labour politician of old, Nye Bevan, why look in the crystal ball when you can read the book?
In this case, the book - literally speaking - is (forgive the plug: I promise that's not the point) one wot I wrote on, of all things The Conservative Party since 1945. More precisely a footnote buried on p. 324, in which I recall what happened after the Attlee Labour government, repealed legislation passed by a Tory government in the wake of the general strike which obliged union members to (yes, you guessed it) opt-in rather than opt-out.
Immediately before the 1927 Trade Disputes Act was repealed by the post-war Labour government, only 22 per cent of the huge AEU (Amalgamated Engineering Union), which had 822,000 members in 1947, paid the political levy. After it was passed, however, the figure shot up to 84 per cent. The figures for other big unions were just as impressive. For the ETU (Electrical trades), they were 31 per cent before and 77 per cent afterwards and for the mighty TGWU, 53 per cent and 99 per cent.
If, then, we take the average 'before' percentage of these three 'super-unions' of their day, then, we arrive at a figure of 35% of members who were willing to cough up consciously in support of 'their party'. Clearly today's trade unionists live in much less class-conscious times. And, for all that they're fighting government cuts, they cannot match the sense of solidarity brought about by the war and its tight labour-market). So no-one is arguing that a third of them will be opting in to Labour, even if that's what it takes to help the Party.
Still, those figures should perhaps provide the Labour leadership with some cause for optimism. However, the fact that there remains a huge gap between the best- and the worst-case scenarios only serves to underline quite how brave - and quite how 'brave' - Mr Miliband is being.