Thursday 29 May 2014

Statistical models and the 2015 General Election


So now we've got lots and lots of polls. We've got a massive dataset from the local and European elections. And we've got about eleven months left until the General Election. The result should be resolving in front of us, based on statistical modelling of past experience, polling trends and electoral data. Right? Well, hold your horses. We're getting a better idea. We have the range of likely results. We're able to exclude, historically, what we definitely think won't happen. And experts' prognostications are beginning to converge, as they usually do. But we're not quite there yet.

So let's take a look at the range of easily-available statistical models - and what they tell us about the general election. We're going to leave the Liberal Democrats out of things for a moment, because we're focused on which of the 'big two' is going to be the largest party. And, let's be honest, it's best to leave them to their private grief for now.

Leo Barasi's blog, using poll data one year out from General Elections: Conservatives 283 seats; Labour 323 seats.
Rod Crosby's simple regression calculation, based on last week's council polls (implied): Conservatives 326; Labour 262.
Stephen Fisher's model, based on the polls' behaviour in the run-up to previous elections: Conservative 307; Labour 284.
Chris Prosser's approach, looking at the last local election before the big one: Conservatives 304; Labour 279.

Oh, you want an average? Well, don't we all. The Barasi and Crosby approaches are the crudest, and probably best to rate lower down the ladder of any overall look at these forecasts. And averaging a load of shots in the dark doesn't mean all that much. So this is basically statistical rubbish. But a mean score of all those seat counts? That'd give the Conservatives 305 members of the House of Commons, and Labour 287. That's a pretty tight, but still a clear, judgement in anyone's language. What happens next will depend on what the Liberal Democrats think about going on in a Conservative-domianted government.

So here's the rub: Labour are unlikely to win the 2015 General Election, at least outright. Their leader's ratings are pretty poor, while David Cameron's (above) are mediocre to okay-ish, and modelling past elections (opens as PDF) on the basis of 'satisfaction with the Prime Minister' gives us pretty strong results.

But it could be close in terms of seats. And Labour still has at least a stab at being the largest party. Less than 20 seats may divide the parties in total. We've been saying this for months, of course - perhaps years. But it's good to have it all parcelled up and quantified. It's a comfort blanket of (apparent) mathematical certainty. And it shows that, on the basis of past elections, Mr Cameron's probably going to be with us for some time yet. But that 'probably' hides a wealth of uncertainty.

Monday 26 May 2014

First thoughts on the European elections


So the United Kingdom Independence Party won the European elections (above). So far so predictable. Though there were senior Labour figures who thought that Labour might nick the contest at the start, there were few of them left by its end.

But what does it mean for the 2015 General Election? Well, start with this - Labour has done quite badly, though by no means as badly as doom-mongering backbenchers and excitable journalists would have you believe. It did make some progress, though it was inevitably held back firstly by the way the contest became a family row within the broader conservative (and Conservative) movement, and secondly by its leader's last-minute campaign gaffes. Ask yourself, though: how good would you look if you were photographed eating a bacon sandwich?

Anyway. Labour beat back a fiercer-than-expected Conservative challenge for second place, mainly with the help of a very strong vote in London. Wherever there were large, cohesive, older, male groups of white voters with a less-than-college education, Ukip triumphed and Labour retreated; anywhere highly-educated, diverse, multicultural and young, Ukip got nowhere, and Labour moved forward (hello, Oxford and Cambridge). London, being pretty much the definition of young, diverse and multicultural, and containing lots of highly-educated voters alongside many who are not, put a shiny gloss on Labour's results - just as it had in the local council elections.

But don't be fooled. The Conservatives may have come third in a national election for the first time in their history, but Labour ended up only between one and two points ahead of them. Based on historical experience in past electoral cycles, that will probably see David Cameron returned as Prime Minister next year, though (as we've been saying all along here at 'Public Policy and the Past') as the leader of a party still without a majority. Say for the sake of argument that Labour gets 32 or 33 per cent of the vote next May, while Mr Cameron leads the Conservatives to the 37 per cent that precedent based on last night suggests he might. Then the Conservatives will have almost exactly the number of seats they have now.

That's where we've been for a long time. It's where the detailed polling on Ukip supporters' likely allegiances in a General Election suggests we'll stand when the dust settles in about eleven months' time: if we just add the likely Ukip returnees to the Conservatives' and to Labour's scores, the former party would again be a little ahead, though both would be some way below the levels suggested by statistical modelling and the current opionion polls.

So the whole thing's been an exercise in confirming our previous impressions: Labour still a little ahead, the Conservatives gradually catching up and in the process of overhauling them, and a likely unhappy return to No. 10 for Mr Cameron, buttressed by fewer Liberal Democrat allies (which is another story altogether).

But it's nice to get it all confirmed in a great big opinion poll involving actual ballot papers in big black boxes. Isn't it?

Friday 23 May 2014

Vote 2014: smash your swingometers


For many years after the Second World War, you could represent British politics as a linear to-and-fro on a swingometer (above). You can see it in action here, from the first 1974 General Election (when Labour did much better than forecast). Sometimes Labour would be on top. Sometimes there would be a surge in the Conservative vote. There were 'floating voters' in between, who'd often vote in a few Liberals too, but in general you were Labour or you were Conservative, and Britons who were prepared to move between those two parties were electoral gold dust.

Now? Not so much. For the (very) preliminary results from last night's local council elections in England gives us a much more complicated picture. Forget all the journalistic fluff about a 'breakthrough' for, or an 'earthquake' involving, the United Kingdom Independence Party. Yes, their performance was remarkable - especially in turning votes into actual seats in council chambers. Everywhere there are lots of older male voters, whether they are prosperous post-mortgagees or poorer council estate residents, that party has surged ahead. But that's not a winning electoral coalition: Ukip are still decisively the fourth party in terms of seats and councils run.

The more important thing to look at, at least as far as the Westminster election that's just over eleven months away, is what Ukip's rise is doing to the electoral landscape. It's distorting and twisting it, so like it looks like one of those puzzles for the eye in which you can never tell where the staircases all end. In London, Labour has done very well indeed, picking up Redbridge, Merton, Hammersmith and Fulham, and the like. Outside, especially in its northern heartlands.

Let's zoom on a case study, shall we? We like those here. In Essex, Labour retreated a bit in the swing Parliamentary seat of Harlow, in the face of a strong Ukip challenge. And they lost control of Thurrock, a must-win seat in 2015. But the Conservatives have got a big old headache here, too, losing control of Castle Point, Basildon and Southend - all clear demonstrations of what happens when right-wing voters defect and split the vote at 'their' end of the ideological spectrum. Doubtless we could give hundreds more examples.

What it all means so far is this: the age of two-, three-, or even four-party politics is over forever. What we're going to be looking at is a kaleidoscopic and balkanised electoral landscape. Labour and the Conservatives are going to face off in the inner suburbs of the big cities. There are going to be Liberal Democrat 'fortresses' (or, more realistically, holdouts) where they are going to cling on, and where they may prove impossible to shift in any circumstances (hello, Eastleigh and Sutton). Ukip are going to fight for the far suburbs and some rural areas with the Conservatives, and in white working-class estates with Labour. The Greens are going to continue to do quite well in urban metropolitan areas. In Scotland and Wales, Nationalist parties are going to be a potentially decisive presence - in Scotland, probably winning a lot of ex-Liberal Democrat voters in the Highlands and Islands when we come to the General Election next year.

For political anoraks, statisicians, numerical modellers, psephologists? A dream. For everyone else? A headache.

Tuesday 20 May 2014

What's behind Labour's poll gloom?


There's no doubt about it: the UK Labour Party's polling is pretty poor. Not really poor, not so bad that we can write them off at the next General Election (at least in terms of having most seats), but their performance is underwhelming for the main party of Opposition, faced with not one but two governing parties who always get the blame for 'events', and presented with an open policy goal almost every day of the week.

The polls can be pretty difficult to interpret at the moment, leading to commentators and observers throwing up their hands and saying 'oh, they're all over the place', or pointing to methodological differences between the pollsters. The two main parties seem to swap or share places in the lead, at least for Westminster elections, all the time. But there's really no mystery. Labour's ratings have been falling since the summer of 2012, when they plateaued in the low-40s, before coming to rest in the high 30s for most of last year, and now dropping into the mid-30s at the moment. There's been little or no sign of a matching Conservative recovery, with at best a gentle upward trajectory noticeable since the start of this year - a fact for which Labour can be grateful, and perhaps can be accounted for by defections to the United Kingdom Independence Party (Ukip).

And Labour's numbers are deflating everywhere, like air being steadily let out of a balloon - only today, news came from Wales that Labour's numbers were down there, too, dialling hopes of six to eight Parliamentary gains down to five. That's pretty good (Cardiff North and Cardiff Central are as close to racing certainties that you're going to get, come election night), but not spectacular. Given that we know that the pendulum usually, but not always, swings back towards governments in the run-up to polling day, it's a sign that Labour's going to have to go some to be the largest party in the House of Commons.

Are they going to do it? It seems less likely now than it once did. Statistical analyses building in previous governments' pre-poll recoveries suggest that the Conservatives are now gradually, very gradually, crawling into a position where they might even claim a tiny overall majority (though this remains pretty much at the outer horizons of their ambitions).

Just as Labour's polling decline has been gradual, and not the sudden wake-up call that the Conservatives' new polling leads might suggest, so its causes are also deep-seated and structural. We've been tracking them for about a year here at Public Policy and the Past, and not much can be done to change the battlefield now.

The main element? The economy is recovering, very rapidly. It's that simple. There's been a massive 34-point leap towards the coalition when voters are asked whether they're managing the economy well or badly - and the Chancellor is more popular than either the Prime Minister or Ed Miliband (above), Labour's leader. Then there's Labour's leader. He's quite a good leader, really - getting all the major calls right, making a coherent case about 'predator capitalism', attacking unpopular payday lenders and train operating companies. Certainly not many Oppositions have been anywhere near dreaming about No. 10 just one term into the wilderness, and Labour still has a shot at returning to power next year. But Mr Miliband's main problem is that the public just don't warm to him - a problem that's getting worse, and not better. Voters may have preferred Prime Ministers to their opponents before - they liked Jim Callaghan more than Mrs Thatcher in 1979 - but never by such a margin, and never in conjunction with such dire assessments of the Opposition's economic agenda. Then, thirdly, there's Labour's problems with its grass roots. Put simply, Ukip (and, at a time of European elections fought under proportional voting) the Greens are beginning to eat into their support base.

Put it all together, and Labour are going to struggle to stay in the mid-30s. They're going to make some advances in May 2015, but will they be enough? Right now, it seems doubtful - and a government that brought you the chaos of Universal Credit, a botched reorganisation of the National Health Service and an economic Plan A that quietly morphed into a Plan B they could have kept to the whole time will sail on regardless.

It's a rum old world, politics.

Monday 19 May 2014

House prices: the Bank of England must act now


The weekend warning on house prices issed by Bank of England Governor Mark Carney (above) is welcome, but overdue. The UK's sudden switchback - from eternal economic winter to fevered boom - is no surprise to economic historians, of course. But it still feels like a handbrake turn-style squeal of gears and brakes, and it's disconcerted policymakers today just as those sharp angles have in the past. The Bank is being left scrambling to catch up.

To use Guardian economics editor Larry Elliott's phrase, the Governor has chosen to ramp up to the 'hot' bits of Defcon Four - and move towards Defcon Three. In Cold War parlance, that means we're half ways towards using the nukes.In this case, that'd be a really sharp rise in interest rates during 2015 - something that might well leave a lot of people looking at a lot of bricks and wondering 'how could I have been so stupid?'

But something has to be done, and soon. Start with this: the housing market always ends up stronger in British booms, and always performs more bullishly, than policymakers think. It's that insight that historians can bring to the debate. House prices 'outperformed' the fundamentals, and all forecasts, in 1963-66. In 1971-73. In 1985-89. And then, most catastrophically of all, from about 2001 to 2007. Take a chart of house price predictions in all of those booms, and then plot reality against them, and the 'real' line is always higher and steeper than what all the economists and statisticians thought at the time. The British obsession with house prices is no longer susceptible to supply-and-demand analysis. Always very difficult to predict, dependent on the amount of credit in the system and very sensitive to institutional rigidities and biases, British house prices are not now a matter of normal economics. They require experts in psychology, social pathology and anthropology. As terrified thirtysomething couples huddle in dingy flat doorways in South London, whispering about whether they can get their parents to stump up another £50,000 to pay for a £450,000 flat that is worth ten times the couples' joint earnings, you need to speak the language mentalities, manias, crashes and busts.

Now it's important not to get too carried away. This is really a phenomenon that's making London and the South-East uninhabitable. Have a play around with the house price calculator here, and most people with a job in the household can afford to rent or buy across most of the UK. But that doesn't help you much if you've just been offered a really good job in London, does it? It reduces labour mobility, cuts labour productivity and gives the whole crazy system one more twist around. And the ripples are indeed beginning to spread out - first to cities such as Bristol, Oxford and Cambridge, and now more widely.

But what can the Bank do? Regular readers will know that 'Public Policy and the Past' has been warning about house prices for some time. And the only real remedy is for more housebuilding. Much more housebuilding. Everywhere - including on 'green belts' that are often sterile and played-out bits of field and hillside anyway. We don't just need an increase in building - we need a step-change that would double levels of bricks-and-mortar output within a Parliament. That won't happen, by the way, but it's still the right answer.

In the short term? The Bank must get much more aggressive. No-one listens to public policy makers any more, and buyers are unlikely to take heed of the Bank's mere words. After the European Exchange Rate Mechanism, Iraq, ID cards, Universal Credit and Andrew Lansley's scrambling of the NHS, why should anyone listen to what officials say? I know that characterisation is unfair, by the way, but voter's disgust at the main three parties shows us that this is what they often think.

Here's a shopping list, in no particular order. The Government's 'Help to Buy' scheme should be ended immediately, buried and forgotten about - though paring it back to help lower-income families, as suggested by Liberal Democrat leader Nick Clegg, would be a start. The Bank could insist on new evidence and new rules for the mortgage checks the UK has just introduced. It's still too easy to get around these for existing customers, and for borrowers with perhaps shaky evidence of their incomes. Stress tests for mortgagees could be made tougher, so that banks would look at buyers' exposure at ten or twelve per cent mortgage rates. And then actual caps could be imposed on prices-to-earnings ratios.

That might hold the line for a while. But make no mistake - the Bank is going to have to raise interest rates, sooner than the Monetary Policy Committee wants to, and probably in the early New Year of 2015. Then, and only then, will we see whether they have the guts to take on the property fever that's gripping the South-East of England - making the richer much, much richer than they would otherwise be, and threatening the economic recovery in the rest of the UK.

For now? Talk is fine, and a useful indicator of where we're going. But the Bank has new powers for a reason. It should use them - not tomorrow, not next week, but today. Now.

Tuesday 13 May 2014

Has Labour's polling bubble finally burst?


The most significant UK political news overnight is the publication of opinion polls putting Britain's main governing party, the Conservatives, ahead of the rival Labour Party.

Polling gurus and statisticians have got used to Labour leads for more than two years. Every time the Conservatives got close, there'd be a flurry of excitement. Then they'd drop back again. It did seem as if Labour had effected a semi-permanent realignment of the British electoral landscape, with a chunk of left-leaning Liberal Democrats defecting to their banner and ensuring that they always polled in the mid-30s - at least.

But now we know that's not true. Both of yesterday's telephone polls - which tend to be the most accurate - had the Conservatives two percentage points ahead. You can quibble all you like with methodology - the poll commissioned by Lord Ashcroft does have a suspiciously large number of non-respondents, while the accompanying numbers when ICM asked about the forthcoming Euro-elections look like a massive change on their last poll. But ICM has always been regarded as the pollsters' 'gold standard', and the most likely to get the final result right. These numbers can't be wished away, much as Labour's incoming strategy guru, David Axelrod, will want to see them in the context of the wider polling landscape, which points at the moment to a virtual dead heat. But you know what? Regular readers will know that any more general view of the electoral landscape has shown us Labour's electoral incapacity for years. Remember their weak performance in the 2011 local and Scottish Parliament elections, only slightly mitigated by a better outcome a year later? This blog decided about a year ago that 'Labour probably can't win the next election... outright. But it still has an evens (or better) shot at emerging as the largest party'.

Now things look even bleaker for Her Majesty's Opposition. The more detailed numbers we have have seen George Osborne's figures as Chancellor continue to rise. The economic recovery is at last being felt in a political spring for the Conservative Party - one that seems quite likely to sustain them in power for quite some time to come. The problem that Leader of the Opposition Ed Miliband (above) has is that he does not seem to have a synthesising, coherent macroeconomic message. Attacking deficit reduction that was 'too far and too fast' was right - and popular. Bashing train companies, energy providers and landlords is pretty likely to get voters nodding too. But what does it all add up to? What is the big idea? No-one seems quite clear, quite yet. They're going to have to get clearer, or Labour risks doing just as badly as they did in 2010.

That's right - the election of the apparently 'derided' and 'hated' Gordon Brown. For too long Labour Party strategists have assumed that they had a base or bedrock of 29 or 30 per cent that they could then build on with Liberal Democrats, younger voters and previous non-voters. They were wrong. Gordon Brown and Alistair Darling were tough, experienced, familiar figures who seemed, in 2010, to have brought the UK through the worst economic crisis since the 1930s. Sure, they weren't that popular, but their numbers were often better than those of Mr Miliband and his Shadow Chancellor, Ed Balls. Mr Brown was rated as doing a 'bad' rather than a 'good' job by a net 24 of voters on the eve of the last General Election: Mr Miliband's rating right now is -25. Labour planners appear to think that their numbers just have to go up, a dangerous illusion that envisages the electorate as a series of segments or blocs that can be moved around at will. In fact, things are much more fluid than this, and Labour could nearly as easily go backwards as forwards in 2015.

We've tended to rely, up to now, on statistical models that assume that a government will advance in the year before a General Election, and that the main Opposition Party will fall back. That has tended to produce models suggesting a very tight contest next May, with both Labour and the Conservatives having a chance of being the largest party. But the models are of uncertain veracity this time (which is why it's good that their authors publish and test them, to tell us what's the same as before, and what's changed). The Conservatives haven't been moving forward. It's Labour that's been falling back. A deep dive into yesterday's data shows that their latest precipitate falls are partly due (opens as PDF) to 'their' voters moving over to the United Kingdom Independence Party. If we're allowed to speculate for a bit, UKIP polling at ten per cent or so might hurt the Conservatives; up in the mid-teens, its support might start to eat into Labour's scores as well. If this is right, once the UKIP wave of the Euro-elections is over, we might go back to some Labour leads. But there'll be small, and very vulnerable to the Government's continued trumpeting of the (very rapid and vigorous) economic recovery.

The rise of UKIP, and the general mood of disenchantment with all politics and politicians, is changing all the rules. The Government isn't much liked. Now we know that Labour's support is very patchy and very soft too. What that means is that it is now pretty likely that Mr Cameron will be able to continue in No. 10 after the next General Election, albeit as a minority Prime Minister. But one has to ask: who would want to?

Friday 9 May 2014

Electoral geography could deny David Cameron the majority he craves


As the opinion polls tighten and the Conservative Party gets closer and closer to the main Opposition Labour Party's score, it becomes more and more likely that Prime Minister David Cameron will get another five years at No. 10 Downing Street. But he's unlikely to get there on his own, and he'll again need the help of the Liberal Democrats or smaller parties.

Why? Because the UK's rapidly accelerating economic growth is spread very unevenly - mainly in Scotland, London and some parts of the South and East. The jobs recovery is even more thinly spread, with London taking the lion's share of the improvement in the labour market. An unstable house price bubble, a run-down in the savings rate and just the beginnings of an industrial and investment pick-up are a witches' brew, unlikely to last for all that long. But they'll take the present government over the line a year from now, in the General Election of May 2015, so that's all we need to know for today.

The electoral impact is fascinating. It has led Labour down the dangerous road of ruthless targeting - of rail commuters (hello, Reading West) and privately renting tenants, both clustered in a series of winnable seats - in a recognisable but warmed-over parody of the Obama campaign's data-driven precision.

But the implications are more dangerous for the Conservatives. Organisationally and culturally, their roots have withered north of Birmingham and west of Swindon. They are clinging to outposts beyond those boundaries, even as they are led by a relatively popular and relatively plausible front man in Mr Cameron. What Welsh, Scottish and northern English voters would make of Boris Johnson as Conservative leader, I'll leave to your imagination.

The latest polls we have of northern English marginals show that the Conservatives will struggle to hold on to most of what (rather paltry) gains they made in the 2010 election. Stockton South, for instance, with a large Liberal Democrat vote for Labour to squeeze as they close in on the Conservatives' majority, must already be gone. All right, the Conservatives can hope to capture Berwick, especially as the long-standing Liberal Democrat MP, Alan Beith, is going to stand down in 2015. But insiders rightly believe that the Liberal Democrats are going to be much harder to dislodge than their national poll scores imply, partly because Labour-learning voters will assist them in holding on.

Then let's have a look at Wales, where the latest opinion polls imply that the Conservatives will lose two or three of the eight seats that they currently return MPs from the principality - so they'd be down to just five or six there. They might as well not bother with liberal, relatively affluent Cardiff North (majority: 194) for instance, and spend their money elsewhere. And Scotland? Okay, they might once more well be able to return one single MP, although if the Liberal Democrat vote collapses (and the last Scottish Parliament vote suggests that it might) Labour have a good chance of wiping them off the Scottish electoral map altogether.

It's a doleful picture, really - a Conservative-dominated government returned, possibly with a very similar number of MPs, but further weakened outside of southern and eastern England. A more disunited kingdom than ever, led by a party possibly propped up by one of Westminster's regional micro-parties. That's the thing about the maldistribution of wealth, opportunity and jobs that we've let ourselves be saddled with: sooner or later it was always bound to come back and bite those who helped cause it.

Monday 5 May 2014

What about the times when there's too much history?


So we know that we've been moaning a bit recently about two related problems of the historian's craft: what about the times when you don't have enough time and research cash to bring home the archival bacon (or at least enough of it for comfort)? What about those trips that don't turn up as much as you thought, or perhaps give you a load of questions far from your original brief? So far, so obvious, our answer being really: do your best. Apply your training. Tell the reader about the questions you have. Be realistic. And - no doubt - a hundred other platitudes as well.

But it's just as important to note that the opposite problem can often be just as bad. What about those topics that are just too rich for description? When the archival record is just too big? This historian once got asked to take on a historical project that amounted to two or three huge warehouses. Faced with such a challenge, you will be unsurprised to hear that I fled the scene, fearing for anyone's career and sanity if they had to try to climb that mountain of paper.

It's in this connection that a passage from James Boswell's 1773 journals on his Scottish travels with that famous man of letters, Dr Samuel Johnson (above), struck us with particular force. We've been trying to chug through this in our spare time, which is a bit of a mixed blessing. But do have a read of just two edited paragraphs right next to each other on the page, if you've got a mind that's ready to be daunted by some tangled implications:
This was a good day. Dr Johnson told us at breakdfast, that he rode harder at a fox chase than any body. 'The English', said he, 'are the only nation who ride hard a-hunting. A Frenchman goes out upon a managed horse, and capers in the field, and no more thinks of leaping a hedge than of mounting a breach...Our money being nearly exhausted, [after breakfast] we sent a bill for thirty pounds, drawn on Sir Wiliam Forbes and Co. to Lochbraccadale, but our messenger found it very difficult to procure cash for it; at length, however, he got s value from the amster of a vessel which was to carry away from emigrants. Ther is a great scarcity of specie in Sky[e]. Mr M'Queen said he had the utmost difficulty to pay his servants' wages, or to pay for any little thing which he has to buy. The rents are paid in bills, which the drovers give... If there were encourgement given to fisheries and manufacturers, there might be a circulation of money introduced. I got one-and-twenty shillings in silver at Portree, which was thought a wonderful store.
One can only say: wow. Here's such a lot of stuff that Johnson himself might have spent days decoding it. Some racial prejudice, and some national pride; some history of horseriding and of blood sports; some good material on the lack of a coin-and-note economy in the West of Scotland, late in the eighteenth century; a sighting of an emigrant ship's captain, just as the great migration was getting underway; a good line of potential research on the different amounts you might get for your bills across the country; and, finally, an insight into the hard times the inhabitants were having; the way the economy actually worked, in a rough-and-ready bartering sort of way.

It's all reminiscent of that 'thick description' I was taught of as an undergraduate, a set of insights often illustrated by that old-but-good example, the Balinese cock fight (opens as PDF) that is supposed to tell us much about the society in which it took place. Whatever the pluses and minuses of that approach (and the latter are many), 'thick description' is exactly what we need here: what Clifford Geertz once called 'the symbolic dimensions of social action - art, religion, ideology, science, law, morality, common sense'. Boswell and Johnson's travels exhibit quite a few of those, all right.

So our problems are not only that we don't have enough time and money. That sometimes our archival visits take us down blind (or new) alleys. But historians are also assailed by a tide of material and meaning, equipped with tools that are often incompatible with other historians' understandings. Here the best approach seems to be to unpack the layers of evidence and meaning, one by one, like peeling an onion. Elsewhere numerical or sociological approaches might be better. Only one thing is for sure: time, money and practicalities are not our only enemies.