Taxing preferences and immobility

Note: All the tax rates and allowances in this article refer to the situation in England. Marginal tax rates are typically higher in Scotland, but I don’t want to keep repeating this qualification. However, this means that the disincentives for regional mobility within the UK are even greater for moving to Scotland than to other regions.

One of the difficulties in having a reasonable discussion of tax policy, especially with respect to its impacts on the distribution of income and wealth, is that most policymakers and commentators rely on comic book versions of how taxpayers respond to taxes. The analysis of tax incidence is complex and, most important, it is crucial to distinguish between short term effects and longer-term responses.

In some commentaries you may encounter reference to behavioural changes, but do the authors have any empirical understanding of what this means? I will start with an apparently simple question: Why are both junior (or trainee) doctors and senior (Consultant) hospital doctors so unhappy about their pay? After all, they are employed in a profession that is, traditionally, well regarded and well rewarded, though it may be very stressful. The disadvantage, in the UK at least, is that the profession involves working directly or indirectly for employers which tend to be the epitome of an inefficient and incompetent bureaucracy.

The pay for a mid-level Senior Registrar is about £55,000 per year, in the top quartile of annual salaries for employees aged 30-39.[1] This is considerably higher than the average salary of £45,000 for a university lecturer, most of whom have spent longer in education and were not paid as well during their post-education training. However, as their seniority and experience increases many junior doctors find themselves facing sharp increases in their marginal tax rates.

The 40% rate of tax starts at a salary of just over £50,000 per year. Employee pension contributions are nearly 11% above £51,000 per year, while student loan payments and employee national insurance take an additional 11% of any salary increase. Overall, at this level a junior doctor faces a marginal tax rate of 62% and that is before quirks in the tax system relating to child benefit.

In gross terms, junior doctors are not badly paid and can expect to earn much more once they qualify for Consultant posts. However, over the last two decades the overall structure of the tax system has become more complex and progressive. When the complexity is taken in account, effective marginal tax rates of 65% to 75% are not uncommon.

It is not surprising that many junior doctors are unhappy with the situation, because their net pay relative to both senior doctors and those in the same age group but in other professions has deteriorated significantly over the last two decades. On the other hand, it is difficult for hospital employers and the NHS more generally to address this unhappiness. To give an increase in take-home pay of £1,000 per year to a Senior Registrar will cost a hospital about £3,660 once employer’s national insurance and pension costs are included. Of course, most of this cost is simply a transfer from one government account to several others, but in budgeting terms the wedge between employer costs and employee take-home pay is enormous.

The marginal tax rates for hospital Consultants are even higher because they are affected by the progressive reduction in the personal allowance for those whose net income exceeds £100,000. For newly promoted hospital Consultants their marginal tax rate is over 83%, which should fall to about 60% after their student loans are paid off and they have acquired sufficient years of service as a Consultant. Then, with a bit more experience limits on total pension contributions kick in along with higher tax rates, so that their marginal tax rate can exceed 100%.

The tax system as it affects hospital doctors is simply mad. It is the product of a bureaucratic and political system in which no one is able - or interested – to take a coherent overview of the impact of decisions about taxation as well as financing pensions and higher education. Decisions about each component are made according to a mixture of immediate fiscal pressures and almost complete ignorance of their cumulative impact.

There is another key element, what I will call taxing preferences and immobility. Tax officials are likely to ask about any proposal which might cause or increase such high marginal tax rates: What are the hospital doctors affected going to do? How many will move to Australia or Canada or elsewhere, and can they be easily replaced? In the short term, there is no doubt that the state can exercise large monopoly power over the pay of skilled employees in many sectors which it finances.

However, in the longer-term governments consistently under-estimate the extent to which people’s behaviour changes in response to such incentives. Doctors moving abroad are a very small part of such changes. The biggest response is the decision by many doctors to work part-time rather than full-time because, at the margin, the additional after-tax pay from working longer hours is too small. Another adjustment is to retire earlier than doctors in previous generations or to switch – partly or wholly - from working for the NHS to working privately.

In the aggregate, such decisions have consistently sabotaged labour force projections, not merely for doctors but for other medical professions, and explain why the perceived performance of the NHS has deteriorated so much. Relying upon migration to fill medical labour shortages is at best a stopgap and has helped to fuel the broader resentment about immigration and its effects.

Perhaps worst of all, the high marginal tax rates have a dire effect on essential mobility in the medical labour market. Suppose that differences in population growth or early retirements or other factors mean that an NHS trust in, say, the East Midlands would like to recruit experienced Consultants from outside the region. The trust is willing to offer accelerated advancement on the Consultant scale equivalent to £15,000 per year in pre-tax salary, which would translate to an increase in post-tax pay of between £3,000 and £5,000 per year depending on age and personal circumstances.

Now, consider the costs of moving, especially the transaction costs of buying and selling a house. The new employer is likely to cover removal costs but rarely agent’s fees and never transaction taxes. The average price of premium houses in the East Midlands in 2025 was close to £700,000. The cost of SDLT (what used to be called Stamp Duty) on buying a house would be £25,000, while estate agent fees would be close to £12,000. Hence, any hospital Consultant moving from an area with similar house prices to those in the East Midlands would incur moving costs that amount to a minimum of 7 times the net salary increase and might be as large as 12 times the net increase. Unless personal circumstances provide an extremely strong incentive to move, the likelihood of the NHS trust finding someone to accept such an offer is pretty low.

Such calculations may seem harsh, but they are not unrealistic. I have seen their impact when trying to recruit professorial staff for university departments and, indeed, went through a similar exercise when approached by other universities. That was at a time when both marginal tax rates and house prices were lower. It is extraordinarily difficult to persuade established senior staff to move between employers in sectors dominated by state-funded institutions with national pay scales.

The consequence is that progressive tax structures on income and property transactions have completely messed up the functioning of labour markets in what we can broadly define as the public sector. For me this is “déjà vu all over again”. Over 4 decades ago my work was best known among economists and policymakers for having demonstrated the consequences for employment and unemployment at a national level of rigidities in the operation of council housing. It is sad and extremely irritating that careless or incompetent decision-making driven by short term considerations has reinstated such rigidities.

There is a corollary to this analysis that is relevant to current discontents. The point in career progression when there is greatest geographical flexibility is at entry to the labour market. This may be when medical graduates take up training positions or (rather less) on transfer from junior doctor posts to Consultant positions. However, the greatest source of flexibility, especially at the senior level, is immigration. New entrants to the labour market, whether UK nationals returning from abroad or those born and trained outside the UK have become a crucial source of flexibility in labour markets that do not function efficiently because of tax and institutional incentives. Note, for example, that immigrant hires do not face the additional 9% marginal tax rate for student loan repayments.

Then, there is the “oh people won’t really leave” delusion. In the short run that may be true. However, I lived through the last period (in the 1970s) when marginal tax rates reached stupid levels. The cumulative loss to the UK’s universities of a significant fraction of the most able young and middle-aged academics in fields of high demand including economics and sciences was never repaired. The most visible manifestation is the number of British academics at universities in the US who have received Nobel Prizes, often for work that they started in the UK but continued in the US.

Almost every academic or student of real talent had to make a choice about whether to stay in the UK or accept offers to move to the US. Many of my friends, colleagues, and students chose to go. Almost none came back before retirement, so the loss was permanent. At that time, we chose to stay for personal reasons, partly because of my wife’s health and partly because she (an American) didn’t want to bring up a family in the US. That explains why one should treat high marginal tax rates as relying on a combination of preferences and immobility to collect revenue.[2]

Fixing the problem of inflexible and failing institutional labour markets for highly skilled workers is extremely difficult. Both junior and senior doctors lobby for a general increase in real salary levels, but this does nothing to rectify the impact of marginal tax rates on incentives for flexibility. If anything, a general increase will strengthen the incentive to work part-time or, eventually, to retire early. Offering new senior employees salary increments or one-off bonuses may be resented by existing employees and are rarely sufficient to offset the impact of tax disincentives.

The private sector can find more creative ways of dealing with the disincentives created by high marginal tax rates. The most obvious are bonuses, share options and outsourcing. These and other adjustments rely upon some combination of pay flexibility and privileged tax treatment of different kinds of “income”. Such arrangements generally convert fixed salary payments to risky profit- or performance-related payments, which will usually be higher to compensate for the transfer of risk from the employer to the employee.

Such adjustments mean that higher tax rates may not, in the medium and longer term, have a large effect on the distribution of post-tax incomes for private sector workers. Instead, most of the burden falls on employers who must find ways of protecting their higher paid employees from the impact of higher tax rates. As always, such costs will be passed on to customers in the form of higher prices and lower levels of service. Where the scope for passing the costs of higher taxes onto customers is constrained, either by market conditions or regulation, business will either cut back on their activities or focus their efforts on customers who are willing to pay premium prices.

Think of the wide range of self-employed workers – from plumbers and electricians to accountants, artists and programmers. Up to a revenue of about £50,000 after deducting expenses their marginal tax rate is 26%. That marginal tax rate increases to at least 42% above that level and may be over 50% due to the reduction in child benefit payments. Further, depending on the balance between income and expenses, the VAT threshold of a turnover of £90,000 per year will push the effective marginal tax up to 60% for a net revenue more than £60,000 per year.

These numbers imply that if, for example, they charge £40 plus expenses per hour they can work for an average of 24 hours per week and receive a net income of about £30 per hour. However, if they were to work 35 hours per week, their net earnings after tax for additional hours would fall to less than £20 per hour. If a self-employed worker wants to work full-time and earn a minimum post-tax income of £30 per hour, they must charge at least £60 per hour plus expenses. The tax incentives to work part-time are extremely strong.

Stepping back, over the last two decades there has been a concerted effort to shift the tax structure towards higher marginal tax rates (though often hidden) with more of the apparent burden falling on higher income taxpayers. Such changes are represented as “placing more of the tax burden on the broadest shoulders”. In practice, this has been largely a shadow play. Any competent tax analyst (which excludes most policymakers) knows that in the medium and longer term the tax burden gets shifted to either consumers or immobile factors of production. So, you can tax rich people who insist on living in London or Paris despite high taxes, but sooner or later most of them move or get better tax advisers.

However, the issue that is barely recognised by politicians and other policymakers is the strong bias in favour of part-time and flexible work contracts. The potential impact of high marginal tax rates on labour supply has long been recognised. Economists have tended to play it down on the grounds that either people don’t move or jobs are inflexible, so the choice is between working full time or not at all. That has changed as employers, especially in the public sector, are required to allow part-time working in most circumstances. In addition, the tax structure favours both self-employment and contracting arrangements for which flexible working time is a core element.

The evolution of tax policy over this period highlights the short-sighted behaviour of policymakers. Measures are introduced which yield significant revenue initially but whose returns rapidly tail off. Rather than reviewing and withdrawing such measures, new measures are introduced which, again, produce a temporary boost to revenues but which exacerbate the issues created by the initial intervention. The outcome is a tax system that is over-complicated, inefficient in raising tax revenue, and often daft. However, because of the focus on short-term revenue effects, there is strong resistance to making changes that may reduce revenues at the outset, but which might lead to more efficient outcomes in the longer term. Running a large fiscal deficit reinforces that resistance, even when it is generally accepted that the cumulative impact of the tax structure has a strongly negative effect on economic growth as well as tax revenues.

The current pressure for a wealth tax of some kind is a classic example of that behaviour. There is little doubt that a wealth tax could collect significant revenue initially. Many of those subject to such a tax are reluctant or unable to change their lives and tax arrangements quickly, so a wealth tax is a tax on preferences and immobility. However, all the evidence shows that wealth taxes provide what is effectively a once-off boost to revenues. Not only do the targets move or change their behaviour in other ways, but they create a permanent level of distrust and avoidance behaviour that lasts far beyond the initial gain. Once-off wealth levies may have a minimal impact if they are not repeated, but probably no-one who knows the UK’s fiscal position and past behaviour will believe that a once-off levy is truly limited.

This is the trap that the current Chancellor finds herself caught in. To deal with the mess created by the last two decades of tax changes would involve large increases in the base rates of either VAT and/or income tax. Quite apart from the immediate political cost, would anyone affected believe a commitment to simplify and rationalise the tax structure?

The most plausible outcome is further tinkering with the tax structure that will exacerbate the existing mess and reinforce the reasonable assumption that matters will only get worse in future. That was exactly the view in the 1970s which meant that change, when it came, was both painful and strongly resisted by those who had taken advantage of the loopholes created by high and over-elaborate tax structures.

[1] For readers not familiar with the career structure for hospital doctors in the UK, a Senior Registrar is the top level of training position. Provided that they complete their training and pass professional exams, most Senior Registrars will progress to Consultant (senior doctor) posts. The career structure for General Practitioners (GPs) - i.e. non-hospital doctors - is different, but many of the incentive issues that I describe for hospital doctors apply to GPs as well. The complication is that GP practices are usually independent partnerships. Most GPs have a choice taking a simple salaried position (often preferred by those who want to work part-time) or one that will lead to a partnership after some time has elapsed,.

[2] To complete the story I did move to the US eventually, but this was prompted by an offer from the World Bank that was more interesting than remaining in academic life.