Second guessing the pollsters

With polling day less than two weeks away, this curiously flat election that nobody really wants has yet to take off. The opinion polls, for what they are worth, suggest that the Conservatives will easily gain a parliamentary majority as they enjoy a lead of 13 points over Labour. The Lib Dems are struggling to make themselves relevant and the Brexit Party has virtually disappeared as a coherent political force.

It was all so different in the spring, when the Conservatives, Labour, the Lib Dems and the Brexit Party were all capturing a similar share of the vote, around 20% (chart). Whatever else you might say about Boris Johnson, he has the ability to tap into what a lot of people want to hear, and by promising to “get Brexit done” he has consigned Nigel Farage to the political sidelines by taking votes away from the limited company masquerading as a political party. Jeremy Corbyn is unable to generate any form of feelgood factor. As I have long suspected he will not be in any position to repeat his decent performance in 2017 because he has dithered on Brexit, and in the eyes of many voters he is simply untrustworthy. Meanwhile, the Liberal Democrats’ new leader Jo Swinson does not come across well with voters and it increasingly looks as though her party’s commitment to revoke Article 50 was a major tactical blunder.

Whilst the opinion polls can be wrong, it certainly looks as though the opposition parties will have their work cut out to limit the extent of the Tories’ majority. Since the headline polls have proven to be a poor guide to electoral outcomes in the recent past, the commentariat paid a  lot of attention to the release this week of YouGov’s Multilevel Regression and Post-stratification (MRP) model results. This model, which correctly called a hung parliament in 2017 when predictions based on aggregate survey results indicated a large Conservative majority, suggests that on the basis of current data the Tories could win a 68 seat majority on 12 December. I briefly touched on MRP models in the wake of the 2017 election (here) but it is worth reminding ourselves of what political commentators – who would not normally care about regression models – are getting excited about.

The MRP model proceeds in two steps. First, YouGov builds a detailed description of UK local populations to determine the characteristics of each parliamentary constituency. The modellers then use YouGov survey data to determine how voting intentions are associated with individual population characteristics (e.g. how likely a person is to vote Conservative or Labour based on their education levels or their age). Combining these two pieces of information, using survey data from the preceding seven days, allows pollsters to predict voting intentions at the constituency level. It all sounds very scientific but a few points are worth noting. For one thing, a track record based on one set of observations is not very useful. As YouGov themselves note, “despite the strong performance of the method in the 2017 election, it is not magic and there are important limitations to keep in mind.” Second, it does not offer a prediction for what will happen at the election since data may change in the interim. Third, the model is only as reliable as the data input and we can never be sure whether respondents are telling the truth about their voting intentions. Finally, the sample sizes used in each constituency are very small and thus subject to significant sampling error.

Moreover, despite all the work which has gone into constructing the model, it does not generate significantly different results from a simple method which applies a uniform swing to each constituency. Whilst it would be nice to have access to all the data in order to generate an MRP model of my own, it is impossible for individuals to recreate a sample of 100,000 interviews in a short space of time. This got me thinking about whether there are other ways to generate constituency models and I report the results here, albeit subject to huge caveats.

Using Logit models to predict the election outcome

The starting point is to try and find readily available information on a constituency basis that might help us. I start from the premise that MPs who have strong local support and who recorded a solid majority last time out are more likely to be re-elected. Even if the MP is no longer standing for re-election, I assume they enjoy the benefit bequeathed by the previous incumbent. This is proxied by the size of the sitting MP’s majority relative to the total number of votes cast (or alternatively, the share of the vote achieved by the winning candidate). Since Brexit is such an important factor in this election we can also assess whether the constituency’s pro- or anti-Brexit bias is important in determining the outcome (see here for the results collated by Chris Hanretty). A final variable is the regional polling data which, although not available at the local constituency level, is assumed to be representative for each constituency in the region (e.g there are 73 constituencies in London and I assume that support for each party is broadly the same as the London average).

My model is designed to predict whether the incumbent party retains the seat at the 2019 election. To do this, I ran a series of qualitative choice models (technically akin to a Logit model with fixed effects) for each of the five main parties (Tories, Labour, Lib Dems, SNP and Plaid Cymru) across all constituencies in GB (Northern Ireland was excluded). Comparing the results for the five models across constituencies, I looked for the party with the highest probability of winning the seat. The central case forecasts gave the Conservatives between 333 and 351 seats (corresponding Labour figures: between 243 and 220). The SNP took anywhere between 37 and 44 seats in Scotland (though I reckon it could go as high as 50) whilst Plaid Cymru took between 3 and 5 Welsh seats. The model struggled to give the Lib Dems many seats at all. Even making some manual adjustments, it is difficult to see the Lib Dems picking up more than 15 seats.

How do the results compare with YouGov? The answer is pretty well. Their central scenario gives the Tories 359 seats; Labour 211; Lib Dems 13; the SNP 43 and Plaid Cymru 4. For a lot less effort (basically, some playing around with the data in a spreadsheet and a few lines of code in EViews) I can broadly replicate the results. Crucially, the evidence from both models suggests that the Tories can win an outright majority in the December election. As noted above, this is not a done deal by any means – there is a large margin of error associated with any such model. Electoral Calculus, which runs a similar model to YouGov, also looks for the Tories to win 331 seats (Labour 235) but with a range between 252 and 429 (Labour 141 to 304). You might think that is a sufficiently wide margin as to be meaningless, but they do ascribe a 63% probability to the chance of a Tory majority.

Reality does, of course, make fools of us all. But I am satisfied that my low budget modelling exercise replicates the work of the highly-paid pollsters. I can thus either get it right at a much lower cost – or can save someone a lot of money by getting it wrong for a lot less

An ultra-Conservative manifesto

The economics of the manifesto

Last week’s presentation of Labour’s election campaign promises was fizzing with ideas and as Philip Collins wrote of the plans in The Times, “if I were running a radical think tank, with no responsibility for implementing a word of this, I might marvel at my handiwork.” The low-key release of the Conservative manifesto yesterday was the complete opposite. Paul Johnson, Director of the IFS, wrote of the Tory plans, “if a single Budget had contained all these tax and spending proposals we would have been calling it modest. As a blueprint for five years in government the lack of significant policy action is remarkable.”

It really was a content free collection of what I hesitate to call ideas. Of the main economic proposals, the Conservatives promised not to raise the rate of income tax, VAT or National Insurance (NIC). As the Tories learned to their cost in 2017, when plans to raise self-employed NICs caused such an uproar amongst backbench MPs who pointed out that it breached their 2015 election pledge, taking these key levers out of the fiscal equation could severely limit the Chancellor’s room for manoeuvre. It is not good policymaking. The plans also envisage raising NIC thresholds to take lower paid workers out of the tax net. The commitment to raising the threshold to £9500 next year will cost just over £2bn per year and the medium-term aspiration to raise it to £12500 will ultimately cost £8.6bn. Some revenue will be clawed back by Boris Johnson’s pledge not to further cut the rate of corporation tax, which was scheduled to be reduced from 19% to 17% next April, and which will give an additional £4bn of resources in fiscal year 2020-21 compared to the latest OBR revenue projections, rising to £6.2bn by 2022-23. In effect, the government has taken away from corporates to give to low earners which is a nice nod to progression in the tax system.

One thing the Tories have not done is to promise a reduction in the income tax burden for the better off, as Johnson indicated in his leadership bid over the summer. From an economic perspective this is a good move: It is (a) unaffordable and (b) regressive. Against that, it is yet another Johnson pledge that has not been fulfilled. Nor are there any measures to “fix the crisis in social care once and for all” as Johnson promised in his first post-leadership election speech. To do so would be expensive, and the Tory manifesto maintains the fiction of governments over the decades that the UK can continue to deliver high quality public services without raising taxes to pay for them. This is a fiscal fallacy which has dogged the UK for years and could be swept under the carpet so long as the working age population was growing sufficiently rapidly to generate the means to pay for them. But as the population ages, this becomes increasingly difficult and if the UK is serious about restricting immigration in a post-Brexit world, the problem will become even more acute.

An ageing population will obviously mean an increasing amount of resources will have to be devoted to health services and the Conservatives have promised “extra funding for the NHS, with 50,000 more nurses and 50 million more GP surgery appointments a year.” Although these plans were not costed in the manifesto, fullfact.org estimates that an additional 50,000 nurses would result in an additional £2.6bn per year of government outlays.

Of the costs we can estimate more readily, once we add in the £11.7bn of spending announced in the September Spending Round, the manifesto is estimated to add just £2.9bn of additional spending by 2023-24 – a rise of 0.3% – which is a drop in the ocean compared to Labour’s planned increase of almost £83bn. Investment spending amounts to an additional £8bn – again significantly below Labour’s planned boost of £55bn per annum. All told, this is likely to be sufficient to limit the budget deficit below £60bn by FY 2023-24, or around 2.2% of GDP (even an additional £2.6bn of spending to recruit more nurses would not make a material difference to the overall figure).

... and the politics

But as with all manifestos they are more about the politics than the economics. The reason the Tories have no interest in a “once and for all” solution to the social care problem is that the attempt to present a solution in 2017 backfired spectacularly when it became clear that the plans would potentially require households to run down their savings in a way which penalised certain groups at the expense of others. It caused such a furore that the Conservatives have vowed not to repeat the mistake, even if it means saying nothing at all.

The real politics behind the manifesto, however, is that of Brexit. It is, after all, entitled “Get Brexit Done. Unleash Britain’s Potential.” On this point, the wish list is truly mendacious: It repeats the old tropes about “take back control of our laws; take back control of our money” as if the UK had lost these functions as a result of EU membership. The Conservatives also promise that “we will negotiate a trade agreement next year – one that will strengthen our Union – and we will not extend the implementation period beyond December 2020.” Most experts agree that to get the kind of deal that the UK wants in anything less than three years would be a miracle, and a period of five to seven years is more like it.

Recall that if it looks unlikely the trade deal will be signed by end-2020, the UK has to let the EU know by 1 July that it wishes to extend the transition period. By ruling out this option the UK is merely creating another set of unnecessary red lines. Worse still, by refusing to countenance an extension with no guarantee that a deal will be in place (it is nowhere near as “oven ready” as Johnson says it is), the risk of a no-deal Brexit at end-2020 could come back onto the agenda. And if that happens, even the miserably thin set of economic promises in the manifesto will become much harder to deliver.

Labour’s plans are open to criticism for the extent to which they rely on much higher levels of taxation to achieve their goal of redistribution. But the Conservative plans get the same short shrift because they offer no vision of what they want to the UK to be, other than out of the EU, and accordingly offer no economic solutions. If this is, as many commentators claim, the most important election for decades, the main parties are offering the electorate a truly desperate set of choices: More taxes or more austerity, and Brexit to boot. It’s no wonder that many voters find the choice before them unpalatable,

Labouring under an illusion

The release of the Labour Party’s election manifesto today was a big deal. It has been described across much of the mainstream media as representing a swing to the left, proposing a significant increase in the role of the state including a big nationalisation programme and a major increase in government investment. It certainly represents a radical departure from the conventions of British politics over the last forty years by proposing fewer market solutions to economic management issues. As regular readers of this blog will know, I have long argued that the economic model of the past four decades, in which the market is predominant whilst the state plays a subordinate role, has run its course. But the plans presented by the Labour Party arguably represent a swing of the pendulum too far in the other direction.

Whatever one’s views, however, an impressive amount of work has been put into the economic plan. We can roughly divide it into two areas – redistribution and investment. Turning first to the redistributive element, the document outlining Labour’s funding plans is a serious piece of analysis, the likes of which I do not recall seeing in an election manifesto. It entails a significant medium-term increase in current outlays on areas such as education, health and social care and work and pensions. Total current outlays by fiscal year 2023-24 are projected to be £83bn higher than measures announced in all previous fiscal events – a 20% increase over current plans (on the narrow definition of spending outlined in the document - it represents an increase of around 10% in total spending).

In fairness, Labour has gone to great lengths to explain how this increase will be funded. Around a quarter is to be generated from raising the corporate tax rate from its current level of 19% to 26% by FY 2023-24 (bringing in £23.7bn). The next largest chunk comes from raising capital gains tax and dividend taxation in line with income taxation (yielding £14bn), followed by £8.8bn from a financial transactions tax. Of the other headline grabbing items, income taxes will be raised on those earning more than £80,000 per year (roughly three times the average wage) which is expected to yield about £5.4bn of revenue. It is notable too, that the plans attempt to allow for changes in behaviour in response to higher taxes so the figures quoted are net expected yield, rather than simply the gross yield. Like the underlying ethos or not, I thought that the funding plan was an impressive piece of economic analysis that people have put a lot of thought into.

The investment spending side of the plan confirmed the expected boost of £55bn per annum (2.5% of GDP). In effect, Labour intends to borrow only to fund investment with the redistributive element of the plan funded by higher taxes. Consequently, the simulation analysis I recently conducted on Labour’s plans still holds. Assuming that in the medium-term Labour injects £55bn per year into the economy, my analysis suggests this will raise the public deficit from 2.5% of GDP in the baseline to around 3.9% by FY 2023-24 (chart) and raise real GDP by 1.8 percentage points above the baseline. Incidentally, this implies a fiscal multiplier of around 0.75 (i.e. a fiscal boost of one percentage point of GDP increases output by 0.75%) which is not far out of line with estimates produced in an OECD paper in 2016 (and cited in Labour’s document).

In order to assuage market concerns regarding its plans, Labour has proposed a fiscal credibility rule which will (i) eliminate the current budget deficit within five years; (ii) maintain interest payments below 10% of tax revenue and (iii) improve the strength of the government’s balance sheet during the next parliament. Part (i) is, in my view, achievable but (ii) will depend very much on how markets decide to set interest rates. Part (iii) is economically very interesting. Labour recognises that there are significant costs associated with its nationalisation plans but in buying up companies the state also acquires an asset. It thus proposes targets that take into account the net balance sheet position of such transactions. The idea is based on work by the Resolution Foundation (here) and it is a genuine fiscal innovation. There are indeed good reasons for incorporating it into the fiscal framework since the public sector balance sheet is increasingly a tool of macroeconomic management (an approach pioneered by central banks in recent years). 

But just because the plan is interesting and innovative does not mean it is sensible. It is a good old-fashioned soak-the-rich strategy, allied with a plan to tax financial institutions and the corporate sector. Paul Johnson of the well respected Institute for Fiscal Studies said in a radio interview that the manifesto will produce “just about the most punitive corporate tax regime in the world”. It will crucify the City of London, where financial services generate 40% of the surplus on services trade and which in turn offsets a large part (though not all) of the deficit in goods trade. Simply put, the UK will be a far less attractive business location. A combination of this economic plan and Brexit would undo decades of work to improve the UK’s standing as a business-friendly location (although Labour does promise to put any Brexit deal it secures with the EU to a public vote and include an option to remain).

It appears from the latest opinion polls that Labour has little chance of getting sufficiently close to the levers of power to actually implement its plans. But it will move the dial. The electorate has had enough of the austerity forced on them over the last decade and the Conservatives will be forced to respond with a policy which also implies additional public spending. As even the FT’s economics correspondent Chris Giles pointed out in an article today  “taxes cannot be something that other people pay.” If the UK is serious about improving the quality of public services, notably the sacred cow which is the NHS, taxes will have to rise. But everyone has to make some contribution and it is dishonest to suppose that only big companies and the rich should pay the taxes that everyone else benefits from. 

It is right that we have a proper debate about the role of the state in the economy. The benefits of a low tax, light-touch regulation regime worked for a long time but in the wake of 2008 the limits of this system were shown up. It’s just that Labour’s 1970s-style socialism is not the way to go either.

Debt or equity?

One of the reasons offered by market strategists for continuing to buy equities is that the dividend yield on stocks is considerably higher than that on government bonds. It is hard to argue with this. The one-year expected dividend yield on the FTSE100, for example, is currently 4.7% versus 0.7% on a 12-month government security. Assuming that equity values remain broadly stable, it makes perfect sense to buy equities which yield a 400 basis points premium over bonds. Now imagine the choice is between corporate debt and corporate equity. From an investment perspective the same applies. But from an issuer perspective things look very different.

UK A-rated corporate debt trades at around 1.94% - more than 270 bps below the dividend yield on equities. Companies thus have an incentive to issue debt rather than equity in order to cut the amount they have to shell out each year in order to persuade investors to buy into their company. After all, according to the well-known Modigliani-Miller theorem the company’s valuation is indifferent to whether it is financed by debt or equity. To the extent that the dividend yield represents a measure of a company’s profit that is redistributed back to shareholders, there are good reasons why a company might want to reduce it – perhaps to increase the funds available for investment or simply to raise employees pay (or even simply to hike the CEO’s bonus).

There have been suggestions that this is one reason why equity issuance is beginning to dry up. The evidence is not conclusive but latest data from the London Stock Exchange does point to a reduction in equity capital issuance over the past couple of years. Based on annualised data for the first ten months of 2019, we look set for a second consecutive decline in issuance with a figure which is roughly one-third below the average of the past two decades (chart 1).

It is indeed notable that equity investors have not revised down their expected returns on stocks despite the fact that interest rates have fallen to all-time lows. We can derive this from the formulation of the dividend discount model attributed to Myron Gordon, known as the Gordon growth model. Playing around with the formula, we derive the result that the compensation demanded by the market in exchange for holding the asset and carrying the risks depends on the expected dividend yield[1] and the (constant) growth rate assumed for dividends. Since the latter is a constant, the required rate of return is a positive function of the expected dividend yield. The expectation that dividends will remain high has thus conditioned markets to demand ever-higher returns.

My calculations suggest that UK equity investors require a total return of 9.8%, which is the highest since the immediate aftermath of the financial crisis in early 2009 (chart 2). If we subtract the risk-free rate from our estimate of total expected returns, we derive the equity risk premium. On my calculations, this is somewhere in the region of 9% in the UK which is comfortably the highest rate in the 25 years over which I have calculated the data (chart 3). Back in the 1990s, I puzzled over the fact that the ERP was negative and concluded that this was flashing a signal that investors were overly complacent about market risks. This in turn prompted me to be bearish on equity trends long before the markets actually corrected (in truth I was way too early so it is no great boast). We cannot say the same today: It may be the case that Brexit-related uncertainty has prompted investors to demand a higher premium but since it has been trending upwards for the past 20 years, this is not a particularly good explanation.

But markets may still be complacent about risks, as they were 20 years ago, albeit for different reasons. In short, investors appear to expect that dividends will continue to rise. The high level for the ERP is thus a misleading signal based on the fact that expected returns are rising whilst the risk-free rate continues to fall. But investors may one day be wrong about expectations of continually rising dividends. This could certainly come about if companies decide that they are better served by issuing debt rather than equity finance, thus reducing the amount they need to pay out in dividends. Issuers do not appear to have adjusted to the fact that the traditional discount of equity dividend yields relative to bond yields has flipped and is unlikely to revert any time soon.  But company treasurers must be wondering whether now is the time to do what governments are increasingly prepared to do – use the period of low yields to issue lots more debt.

---

[1] The true expected dividend yield is expected dividends relative to the expected price but the Gordon growth model depends on expected dividends relative to the current price which is not quite the same thing. For expository purposes, we nonetheless call this term the expected dividend yield.

As more evidence comes to light ...

Nigel Farage likes to claim that he is speaking for the ordinary citizen and has argued strongly against immigration, suggesting that the studies which show the benefits of immigration to be flawed because they fail to account for the cost of in-work benefits such as tax credits. But right-wing politicians are not the only ones to be concerned about free movement within the EU: One of Jeremy Corbyn’s key union supporters, Unite’s Len McCluskey, suggested this week that Labour should also take a tough line on free movement of workers. But the evidence clearly shows that both are ignoring the economic benefits of free movement to the UK and their opposition to immigration is based on a false premise.

Three months ago HMRC published data which showed the revenue and expenditure contribution of EU citizens to the UK’s tax and benefit system for fiscal 2016-17 which paints a very different picture to many of the claims made during the  EU referendum campaign. The figures show that EU and EEA citizens received around £3.2 billion in child benefit and tax credits. But they paid in £21.3 billion, meaning that they made a net contribution to the UK Exchequer of £18.1 billion. Although this figure does not include all tax and revenue streams, it is a safe bet that this these figures are a fair reflection of the overall fiscal contribution of EU citizens. To put it into context, a figure of £18.1bn is more than the current budget surplus for FY 2016-17 (£17.4bn). Without the fiscal contribution of EU and EEA nationals, the current balance on public finances would have been in deficit. 

Citizens of all EU countries made a positive exchequer contribution, with the largest net sums being paid by French (£2.9bn) and Irish (£2.2bn) nationals, with the Poles chipping in a net £2.1bn. On a per capita basis, the biggest contributors were Danish nationals, who paid an average of £25,090 followed by the French (£24,090, chart). The average European per capita contribution was £7,260 with the overall figure being depressed by the relatively low contributions made by central and eastern European nationals. The figures tell us two things: Most obviously European nationals come to the UK to work, not claim benefit as has often been claimed in the recent past. Second, the distribution of per capita contributions highlights the fact that western European nationals tend to be highly skilled and are therefore high earners whilst those from newer EU member states tend to be lower paid.

Even before the end of free movement, the UK is no longer an attractive destination

For all the fact that EU nationals make a significant contribution to the UK economy, there is clear evidence that fewer of them are entering the UK than was the case three years ago. Although the UK immigration data are subject to significant methodological flaws, they are all we have at present and they show that in the year to March 2019, a net 226k foreign nationals entered the UK compared to 311k in the year to June 2016. The net inflow of EU nationals has slowed to 59k versus 189k in the year prior to the referendum, with net outflows of EU8 citizens as many of the huge wave of migrants from Poland begin to return home. But the real kicker is that in the year to March, a net 219k non-EU nationals entered the UK compared to 171k in the year to June 2016. Recall that this is the part of immigration that the UK government can actually control, so if the electorate really does have a problem with immigration they are going to be mighty dumbfounded by the government’s current policy.

It has long been recognised that there are substantial costs associated with pulling up the drawbridge and the government appears to have quietly dropped its unrealistic pledge to reduce the net number of immigrants below 100k. But it might be too late: Large sections of the UK public sector rely on foreign labour input and there has been a 70% fall in the number of EU citizens coming to the UK to work. In the absence of a sudden surge in Brits willing and able to take on the tasks, these positions will have to be recruited from outside the EU, which may explain why the numbers arriving from outside the EU have increased.

Moreover, one of the flaws with the data is that it includes those coming to the UK to study and who make a net overall contribution to university coffers. Prior to the EU referendum roughly one-sixth of the net immigration figures was attributable to students – there has been a reduction of around 20k in new student numbers over the past three years with anecdotal evidence suggesting that many have been put off by the more hostile climate. This is certainly not a desirable outcome for universities which are seeking to expand their footprint in an increasingly competitive global education market.

Whatever one’s views on immigration, it is clear that a policy of restricting the flow of EU citizens into the UK will have economic consequences. This may only be evident initially at the margin. But it remains one of the potential underrated consequences, whose full effects will only become evident in the fullness of time.