This May, British voters will cast ballots for a smorgasbord of different institutions under a plethora of different systems. Given that politicians of all stripes will make claims of vindication and accusations of failure on the basis of these election results, pollster John Curtice explains what, if anything, the numbers will actually tell us.

It will be the biggest test of the popularity of the parties until the next round of European elections, due to be held in mid-2019. Every voter in the UK will get the chance to cast at least one vote, and some more than one. It would thus seem obvious that the elections being held on 5 May of this year should be regarded as a key test of Jeremy Corbyn’s ability to deliver electoral success for his party – or otherwise. Certainly that is the view being expressed by many of those Labour MPs who are not sympathetic to the ideological stance of their new leader.

Trouble is, the set of elections being held in May is a veritable smorgasbord. They are being held under different electoral systems, are for institutions with very different responsibilities, and will vary in terms of the extent to which voters’ decisions reflect their evaluation of the parties at Westminster. Above all, the headline results, at least, will not necessarily provide an unambiguous indication of what progress, if any, Labour has made since last year’s general election.

An electoral map of many scales

In order to make sense of these elections, we need first of all to stake out the terrain over which the parties will be fighting. In Scotland, Wales and Northern Ireland voters will be voting for members of their own devolved parliaments or assemblies. In Scotland and Wales they will be doing so under a proportional system that entitles them to cast two votes, one for a local member of Scottish parliament (MSP) or assembly member (AM) and one for a regional list, with seats being allocated in such a way that the overall outcome is roughly proportional to the number of list votes each party receives. In Northern Ireland, by contrast, voters will be invited to place their local candidates in rank order of preference, with the seats (six in each constituency) being allocated via the single transferable vote system.

So far, so (relatively) straightforward; much less so, however, in England. In London, voters will be invited to express a first and second preference with regards to who should succeed Boris Johnson as mayor, while similar mayoral contests will also be held in Bristol, Liverpool and Salford. At the same time, voters in the capital will, under a system very similar to those in Scotland and Wales described above, cast two further votes that between them will determine the membership of the body that keeps an eye on the mayor – the London assembly. Meanwhile, outside of London – except in Greater Manchester where the post is being wrapped up into the new mayoralty, for which the first election will take place next year – everyone will be invited to vote for their local police and crime commissioner (PCC), again by expressing a first and second preference. PCC elections (of which there will be 40 in total) will also take place in Wales (but not in Scotland) alongside elections to the Welsh assembly.

That, believe it or not, is not all. Many a local council election will also be held in provincial England. All bar one of the 36 ‘metropolitan districts’ – that is, the large councils in and around such big cities as Birmingham, Liverpool, Leeds, Manchester, Newcastle and Sheffield – will be holding elections. So will 19 of the 55 ‘unitary authorities’ – that is, councils that, like the metropolitan districts, provide all the local services in their area, and which again typically cover relatively urban areas such as Bristol, Hull, Plymouth and Southampton. Furthermore, 70 of the 201 second-tier district councils in England that provide some services such as refuse collection and leisure centres (as opposed to bigger, county councils that deliver education and social services) will also be holding a ballot. Many of these councils that will have elections also cover relatively urban areas, albeit mostly those in the southern half of England, including Cambridge, Gloucester, Ipswich and Oxford.

There is also one other important feature of the arrangements for these local council elections. In most cases just one-third (or in a few instances, one half) of all the seats on a given council will be up for election. Only in 18 councils are all the council seats being contested – in each case as a result of changes to local ward boundaries that in themselves will make interpretation of the results more difficult. This arrangement inevitably limits the scope for reversing past losses and making headline-grabbing gains, even though the local government elections are being held under the winner-takes-all ‘first past the post’ system.

Cherry picking, then and now

All in all, it means that some voters, including everyone in London and in Wales, will be handed no less than three ballot papers – and will be expected to complete them in different ways. The Electoral Commission, local returning officers and the political parties will all have a challenge on their hands with regards to ensuring that voters do not get confused. But in any event, it is probably already apparent by now that, with such a complex picture, there may well be plenty of opportunities for politicians of all parties (and all parts of the Labour party) to cherry-pick the results that put their particular cause in a favourable light, and to ignore or misinterpret those that do not.

That this is the case becomes even clearer when we look at when the seats that will be contested this time were last fought over. Crucially, only a few were last contested in May 2015, when in the general election Labour fell back in terms of seats (but not votes) relative to what had already been a serious defeat five years previously. Consequently, we cannot simply assume that a party that loses seats this May is doing worse than it did last May – or indeed that a party that gains seats is on an upward trajectory.

Rather, most of the seats that are being fought over this May were last contested in the first half of the last parliament – though not all on the same occasion. In Scotland and in Wales (and indeed in Northern Ireland) the last devolved election was held in May 2011. At that point Labour enjoyed a clear, but quite modest lead over the Conservatives in the Britain-wide opinion polls of just 3–4 points, while Ukip was still regarded as but one of many ‘others’. In contrast, the four mayoral positions (including that in London), the London assembly, and most of the seats up for grabs on local councils outside of London, were last contested in May 2012. By this stage, not least as a result of the ‘omnishambles’ that had surrounded George Osborne’s budget a few weeks earlier, Labour’s lead had doubled to 7–8 points, while still no more than the very first signs of UKIP’s rise were in evidence.

Meanwhile, what were the first ever elections for PCCs were not held until November 2012, when an embarrassingly low proportion of the electorate – just 15 per cent – actually bothered to vote. That fate at least should be avoided this time around simply because many voters have a local council election (or in Wales, assembly election) to turn out for too.[1] In any event, UKIP’s rise was in evidence by the time of those first elections, but at some 6–7 per cent of the vote the party was still not the force that it subsequently has proven to be. Labour, meanwhile, were at that point as much as 10 points ahead in the polls.

There are then two crucial features to the political backdrop to this year’s elections. First, most of the contests were previously held when Labour was doing reasonably well in the polls – and indeed, with the exception of the London mayoral and Scottish parliament elections, this was broadly reflected in the party’s performance at the ballot box. This was especially true of the local elections in 2012. The BBC’s projection of these results into a Britain-wide election vote suggested that Labour’s performance was worth 38 per cent of the vote, enough to put the party 7 points ahead of the Conservatives. It was easily the party’s best performance in any of the annual rounds of local elections held during the last parliament.

Consequently, Jeremy Corbyn faces a relatively demanding electoral test at a time when many are looking to see if he ‘fails’. Even if Labour were to enjoy some recovery from its position a year ago, the party would still suffer net losses. Indeed, simply repeating its performance locally in last year’s local elections would see the party lose control of Dudley, Cannock Chase, Crawley, Redditch, Rossendale and Southampton. This set of losses would undoubtedly be regarded by Corbyn’s critics as evidence that he had lost the plot in middle England – but in fact they may simply be an indication that the party was previously just treading water.

And that, at least, appears to be what the party is doing according to the polls – albeit they remain methodologically in flux since their underestimation of Conservative and overestimation of Labour support in last year’s election. According to the three polling companies that have so far been polling on a regular basis in this parliament, Labour is currently averaging 33 per cent, no more than marginally up on the 31 per cent the party secured in the general election (and indeed on the party’s standing in these same polls before Corbyn became leader). The party still trails the Conservatives by six points, much as it did in the ballot box last year.

Could this be Ukip’s day in the sun?

Meanwhile, the second key feature of this year’s elections is that the seats and positions being contested were largely last fought over before Ukip had made much, if any, impression on the electorate. True, the use of the first-past-the-post system in English local elections does not make it easy for Ukip to turn votes into seats, but that did not stop the party from winning over 200 local seats in May 2015 on the same day that it picked up just one MP; nor did it stop them winning more than 150 local seats in May 2014 (when there were far fewer seats at stake). Meanwhile, the use of proportional representation in the London assembly election means that any party can expect to win seats as long as it wins more than 5 per cent of the vote, while in Wales the de facto threshold is some 7–8 per cent.

Much like Labour support, Ukip support has largely remained steady since the general election, despite the fact that the party also has witnessed a measure of infighting and internal disagreement. Polls conducted over the internet tend to find more Ukip supporters than those undertaken by phone (much as internet polls also tend to uncover more people who say they will vote to leave the EU), but between them the polls suggest that the party is still not far adrift of the 13 per cent that it won in the general election. Meanwhile, with the referendum on Britain’s membership of the EU being held just seven weeks after May’s local and devolved elections, we can anticipate that Farage will attempt to capitalise on the efforts of the ‘Leave’ campaign by suggesting that those who wish to leave should vote for the only political party campaigning for a ‘leave’ vote.

Certainly, even though cosmopolitan and ethnically diverse London is not a particularly happy hunting ground for Ukip, on its current poll ratings the party would be expected to win a seat on the London assembly, and would also pick up many more than the seven local council seats it won in 2012. Meanwhile, the election in Wales in particular looks as though it could be deliver a good result for Ukip. Five years ago the party won just under 5 per cent of the vote for the Welsh assembly, and came away empty-handed. Now the most recent poll of voting intentions gives the party 14 per cent, and the prospect of nine assembly seats. Indeed, it could end up being the only party to make net gains in Wales – an outcome that would seriously hamper Labour’s efforts to retain its tenuous single-party control of the assembly (it has exactly half of its 60 seats, and on the current polls could easily lose two or three of them).

A poor reflection: The devolved elections

However, experience suggests that the outcome of devolved elections does not necessarily provide an accurate reflection of the popularity of the parties nationally.[2] Labour has perpetually struggled to perform as well in assembly elections in Wales, which instead have proven relatively fruitful territory for Plaid Cymru. Even under Tony Blair’s leadership of the UK, the party never won more than 30 seats in Wales. However, that – unfortunately for Corbyn – was less the case in 2011 when, in terms of votes at least, the party puts in its best devolved election performance yet. Indeed, at 42 per cent, Labour’s share of the constituency vote in 2011 was rather better than the 37 per cent that the party won in the UK general election last year and the 36 per cent it secured in 2010. So, once again a headline count of seat losses could make it appear that the party is going backwards when it is in fact simply treading water.

The same caveats also apply – in spades – to the devolved election in Scotland. As in Wales, until last year at least, Labour’s performance in devolved elections has always been a pale reflection of its showing in UK general elections. To that must be added the fact that, following on from the independence referendum, Scotland’s electoral politics have come to be dominated by the constitutional question – which puts it on a political trajectory completely different from those of England and Wales. What Corbyn says or does is likely to make little or no difference to the outcome north of the border – and so the result there is likely to prove a poor indicator of his wider popularity or otherwise. Meanwhile, disastrous as the result might have been five years ago, simply treading water and winning the 24 per cent of the vote that the party won last May would again see the party lose seats.

Salvation in London?

Meanwhile, if Wales and Scotland will not necessarily prove an unambiguous guide to Labour’s electoral health under Corbyn, neither, in truth, will London. Yet, unmerited though it may be, London could prove to be the Labour leader’s salvation. In 2012 Boris Johnson won an election he should not have. He secured 44 per cent of the first-preference vote while his party won less than a third of the vote in the London assembly election. He won because many a Londoner was willing to vote for him personally, even though they did not support his party. But, having engineered his return to the Commons, Johnson is now leaving the London stage. With his personal vote probably lost to the Conservatives, the outcome this time is likely to reflect the overall popularity of the parties. Despite all of its travails, Labour won easily in London last year and thus should do so this year too.

If it does, we can expect Corbyn and his allies to trumpet the success of the party’s standard-bearer in the mayoral contest, Sadiq Khan. The result, they will argue, is evidence that Corbyn can reach parts of the electorate that his predecessor, Ed Miliband, could not. In truth, such a result might simply demonstrate, as others may, that the party was treading water – and ultimately it needs to do much more than that. But perhaps what would be a straightforward, high-profile and important success in London would be enough to crowd out any disappointments in more dim and distant parts of the UK, and in elections for institutions such as local councils and police commissioners about which few care, the meaning of which, however, would be far from ambiguous. Perhaps – we will find out.

John Curtice is professor of politics at Strathclyde University and a columnist for Juncture.

This article appears in edition 21.2 of Juncture, IPPR's quarterly journal of politics and ideas, published by Wiley.

[1] However, in general this is more true of urban than of rural England. Thus, in those police authority areas in which local elections are being held in some councils but not others, we could find that the turnout is higher in Labour-voting areas, thereby giving Labour a potential advantage.

[2] This reservation also applies to the PCC elections, in which nearly a quarter of the vote in 2012 was won by independent candidates.