What is the overall shape of healthcare work in Britain?

Most of the healthcare work in Britain is provided by the NHS, the largest single employer in the country that currently employs almost 1.3 million full-time equivalent staff in England. However, this does not mean that healthcare work is centrally directed. Due to the 2012 Lansley reforms, NHS trusts exist in a marketised environment where they are set up as competitors with one another. When this is allied with the already patchwork nature of NHS services, e.g. GP services being independent private providers, it means that the experiences of healthcare work can significantly vary from employer to employer, even within the same geographic area.

There is still a gendered split in many of the roles in the NHS, at least in hospitals. Our two porters that were interviewed reported that, in their Estates and Facilities Department, most porters and work staff were male, which tallies with the Nomis data for January 2021 to December 2021 which recorded a fifth of hospital porters as female. The former psychiatric nurse (who was now an employment advisor) confirmed that in his experience the majority of psychiatric nurses had been male, which contrasts with only 10% of all registered nurses in Britain. Indeed, 76.7% of all NHS staff are women, and it is possible that this has had some historic influence on NHS pay with women’s work typically artificially devalued. Interestingly, according to Nomis data for the annoyingly broad category of ‘Human Health and Social Work Activities’ work recorded a significant proportion of women working part-time. In December 2021 Nomis recorded that only 53.36% of female workers in health and social work worked full time, in contrast to 72.48% of male workers. Although the Nomis data does not apply specifically to the NHS, there is no reason not to assume that the NHS (given the proportion of this sample it will constitute) does not follow this broad pattern. There has been a marginal and slow shift as to the gendered nature of ‘Human Health and Social Work Activities’ in 1996 83.9% of this labour force were female and this has steadily declined to 77.8% in 2021.

There is also a high proportion of immigrant labour. As of March 2021, 14.6% of NHS staff (for whom nationality is known) reported a non-British nationality. These staff represent approximately 200 nationalities, with Filipino, Irish, and Indian being the most represented non-British nationalities. However, this labour is not distributed evenly across the health service. For instance, London has 27% staff reporting a non-British nationality while the North East reports only 7%. Britain does not produce enough skilled labour to support the NHS, hence migrants are integral to its continual functioning. This has led some trusts to explicitly target other countries for recruitment; for instance, one of our interviewees related how Western Sussex Hospitals Trust targeted the Philippines to bolster its complement of Nurses and HCAs. It will be interesting to see if this has had a significant impact on the size of the Filipino community in Worthing. Unfortunately, this will only be possible once the 2021 census data becomes available and can be compared with the 2011 census data.

We would have liked to do more analysis of the ethnic composition of healthcare work; however, much of the data (and our skill at interpreting it) was more limited than the above data. For instance, it is useful to know that in 2021 there were 158,000 Indian people recorded as working in ‘Human Health and Social Work’, compared to only 14,200 Chinese workers, and one would imagine that this is concentrated in particular areas; however, other parts of the data were not particularly helpful. Much of the data was too clumped together, as there was a broad category of ‘Any other Asian Background’ and the arguably vexatiously broad category of ‘Black/African/Caribbean/Black British’. Thus, this is something that is largely missing from this analysis and is one obvious area where we could improve; however, while this may limit the below analysis we do not feel that it invalidates it.

The bulk of healthcare workers (and, therefore, work) are centred in hospitals and community health services. This is where nearly all of our interviewees were drawn from. It would have been interesting to see response from people working in small businesses such as GP surgeries and within the NHS bureaucratic machine (for instance various CCGs); however, the insights into the hospital system are still interesting and worth analysing.

Many of these are a significant source of work for the towns that they are situated in; indeed, many of the trusts are essentially anchor institutions. This becomes evident when analysing the number of people within Travel to Work Areas (TTWA) who work in hospitals. A TTWA is an area from which its population usually commute to a larger town or city to work. Based on Nomis’ 2020 population estimate of people aged 16-69 (defined here as working age), there are 20 TTWA that have over 5% of their working-age population engaged in hospital work. Taunton TTWA, which has the greatest proportion of its working-age population working in hospitals, was as high as 9.5%. This is why the tendency within hospital trusts to centralise in one site while stripping functions out of the other sites is particularly concerning for the communities that they are leaving and the labour they are distancing themselves from. For instance, the Western Sussex Hospitals trust in the Worthing TTWA (which has 6.65% of its working-age population employed in hospitals) has recently merged with the former Brighton Sussex University Hospitals trust to form University Hospitals Sussex. There was concern amongst our interviewees who worked at this site that more functions would be centralised in the larger Brighton site at the Royal Sussex County Hospital.

Within hospitals the work is clearly demarcated between the different professions and roles, for instance: porters, housekeepers, receptionists, radiologists, doctors, nurses, managers etc. This means that the shop floor consists of several different relations to production and levels of autonomy. While there are many clinically qualified staff, there are also many who are not. The most recent figures show that 52.1% of NHS staff are clinically qualified.

For instance, one of our interviewees, a porter, described a relatively high level of autonomy with how they did their job as tasks would come in and be assigned to them while they were on shift, but it was left up to them how precisely they would do the job. Another porter who often worked waste shifts said that they were left alone for the whole day with minimal contact from their manager as long as the waste was disposed of and no departments complained about their waste not being collected.

Most of these staff (with the exception of doctors) are on the Agenda for Change national payscale, which means that their labour is constantly valued against that of their colleagues in different jobs. Agenda for Change bands (1-9) hinge on the notions of skill and responsibility, essentially, how difficult is the job to do and how significant are the things/people that a person is responsible for. However, the proof of ‘skill’ rests on the certification of it; therefore, many skills used in the lower banded jobs are not sufficiently rewarded as they are not certified. Even though the experiences of work are often different from one NHS employer to another, the workplace policies that govern the work and terms and conditions of employment are almost identical. This is because there is one national body, NHS Employers, that is responsible for negotiating national terms and conditions with healthcare staff, and this is also the body responsible for advising trusts on these policies and procedures. Thus, changes won by trade unions in a dispute at one trust are often mirrored across the NHS.

The work is often shaped in a distinctly hierarchical way, with doctors normally at the top because of their clinical status and historic influence as a profession when, before the growth of management, they essentially controlled the means of production of health care at local level. Also plentiful is the rise of managerial functions alongside people’s jobs. Low level managerial functions start as early as Band Three on the Agenda for Change pay scale and can often end up taking up a significant amount of time away from the actual job. Part of the reason for the growth of managerialism is that for many of the professions in the NHS, for instance nursing, increased managerial responsibilities are a way to justify pay and promotion when the basic functions of the job are not sufficiently remunerated. This is a structural flaw that is common across public sector pay in general. Consequently, the progression of many clinicians is one away from clinical duties and towards administration, this contributes to an aggressive managerial culture in the NHS enacted by people who have not been trained for management.

This can be seen by examining the structure of the nursing profession, which is as follows: a starting nurse enters as a Band Five (starting salary £25,654), becomes a Deputy Sister at Band Six (£32,305), and a Sister at Band Seven (£40,056), with administration and managerial responsibilities increasing proportionately. One of our interviewees gave the example of their A/E department which had as many band Six and Seven nurses (who were given minimal managerial responsibilities) as Band Fives. Essentially, accelerated promotion was used as a form of A/E premium. This is an approach that worked for this A/E department but is not a viable approach for most NHS staff seeking pay increases.

As with the rest of the public sector, due to government action, the value of healthcare work is kept artificially low. Indeed, there are early signs of a growing gap between public and private sector pay increases as sections of the private sector adjust to the labour shortages caused by the Covid-19 pandemic. There are chronic labour shortages in the NHS across nearly all grades; however, the price of labour has not substantially increased, because the government refuses to increase most public sector pay. In some instances, this has led to some trusts essentially bidding against each other to try to attract skilled agency staff (such as doctors and nurses). However, while the basic rate for skilled agency staff is attractive (for low-skilled, approximately Band One to Band Three, it is abysmal) this can be deceptive. Agency staff often find that, once other employment benefits are factored in, e.g. holiday pay, pension, and enhancements, they are not necessarily better paid than their substantive colleagues. However, for clinically skilled workers, agency work can give them greater flexibility as to when and where they work.

The other consequence of staffing shortages is that the current staff are taking on more work without being remunerated for the increased workload. This has led to more staff deciding to leave healthcare work altogether. Comparing the annual population survey data for April 2019 to March 2020 and January 2021 to December 2021 on Nomis provides useful insight as to the state of healthcare staffing. The first data set reports 669,000 nurses while the second reports 505,000. Equally, the former data set notes the presence of 18,600 hospital porters and the later records 14,700. However, it should be noted that not all healthcare roles have decreased. The Government will have to do more to train and attract new workers if it wants healthcare work to operate at the capacity it currently does. However, there currently no signs of this being done. Even the apparent success story of the increase in doctors (in April 2019 to March 2020 there were 295,500 medical practitioners, whereas in January 2021 to December 2021 there were 335,000) is deceptive. Contrary to first reading, this does not indicate that the medical profession is currently producing more staff than are leaving the profession. Indeed, the number of medical graduates has stayed relatively consistent from 2015 to 2019, and although the number of IMG (International Medical Graduate) doctors has increased, it is not enough to account for this rise. What did happen between April 2019 and December 2021 is that the General Medical Council used emergency powers to grant temporary licences to nearly 35,000 doctors, some of whom were IMG, but a majority of which had left the profession within the last six years. These licences are due to expire in September 2022, and while it is likely that the IMG doctors will remain in the profession, it is less likely that doctors who had already decided to leave the profession will remain for much longer, particularly when there is an overall growing trend of doctors retiring early.

While some other professions (for instance, Medical Radiographers) also saw an increase in the number of staff, the overall number of health professionals was 1,522,100 in April 2019 to March 2020 and 1,415,200 in January 2021 to December 2021. It will require more than just an emphasis on recruiting and training more doctors to ensure the healthcare sector continues to function, and there is a danger that focussing too much on this profession is detrimental to efforts to maintain the others.

Thus, the shape of healthcare work is one that is concentrated in particular outposts and that is state funded but operates under quasi-market mechanics. Different types of labour are pressed up firmly against each other with some of them able to adequately advance their interests while others cannot. It is governed by a managerialist culture but little central direction. Currently, it looks to be creaking under the weight of demand and a shrinking labour supply. Over 10% of British GDP is spent on the NHS, yet it is not entirely obvious what benefits this unfocussed spending increase has had, certainly not to the staff who do the work.

What is happening with the NHS / private healthcare?

The NHS continues in its marketised structure; however, some of this will change with the government’s new Health and Care Bill as it is removing some of the elements of mandatory competition that were present in the Lansley reforms, but this could be counterbalanced by private companies being part of the new Integrated Care Boards. Although this bill, in and of itself, will not lead to complete privatisation, given it further centralises power in the Office of the Secretary of State for Health and Social Care for one, it could give private enterprise the opportunity to have greater influence over the commissioning and provision of NHS services. A likely outcome of this is that the legislation ‘will allow contracts to be awarded to private providers without proper scrutiny or transparency’ which would provide greater scope for the cronyism that was evident during the pandemic.

The pattern that was evident was that more services were being outsourced. They were still being provided under the banner of the NHS and free at the point of use, but the company running them was separate from the employer. One of our interviewees related how the hospital restaurant had been outsourced to a private company and that the security contract had their trust had gone through three different operators. Prior to the pandemic there was relatively significant media coverage of the growth of NHS subsidiary companies; these enable back-door privatisation and the worsening for employment terms and conditions. These are unlikely to have gone away. Many staff, particularly the less skilled, are in constant danger of being outsourced as the financial pressures of administering health services come to bear. On the other hand, UVW organised a long running strike of cleaners at Great Ormond Street hospital, first winning pay rise and improved conditions, then later ending outsourcing.

It is different for doctors, however. Many hospital consultants will do some form of private practice work alongside their NHS work. This is an arrangement that currently suits the private providers, the NHS, and the consultants. The consultants get the benefit of good terms and conditions of employment within the NHS, while being able to supplement their income. This means that the NHS gets the benefit of being able to pay the consultants less than what their skill-level demands, and the private providers get the benefit of having a flexible labour force that provides specialist services. Moreover, it is important to remember that GP practices, the foundations of the NHS’ primary care, are themselves private businesses and GP partners are private business owners and would neatly fit into the petit-bourgeoisie.
It is unlikely that the NHS will ever be completely privatised (though large chunks of its labour force could be), as there will be some services that private providers will have no interest in providing, instead they will want to be able to pick and choose the most profitable services while leaving the state to run the rest.

How are workers currently organising in healthcare?

One of the distinctive features of workers’ organisation in the NHS is the presence of professional bodies that double up as trade unions. All the qualified professions within the NHS have a professional body that represents them. For instance, doctors have the British Medical Association (BMA), nurses have the Royal College of Nursing (RCN), and physiotherapists have The Chartered Society of Physiotherapy (CSP). These bodies have a complex relationship with trade unionism, as they also feel that they have an obligation to represent the profession and not just their members, which can lead to internal tensions. Consequently, many of these bodies are not affiliated to the TUC, though there are exceptions, e.g. the CSP and Royal College of Midwives (RCM). However, it is probably no coincidence that these two bodies are seen as less medically qualified and represent young professions.

All of the big three general unions (Unite, UNISON, and GMB) organise within the NHS, with UNISON having by far the most members, though Unite and GMB have noticeable strongholds of membership. For instance, Unite is well organised in the scientific staff and GMB is well organised in the ambulance service. Both Unite and GMB are trying to expand their membership and Unite has previously made a concerted effort to recruit doctors (who are not eligible for UNISON membership). However, it is unlikely that this approach will be particularly fruitful due to the incumbent advantage of the BMA and the Hospital Consultants and Specialists Association (HCSA). These two unions were able to bar Unite from access to the national negotiating machinery in the latest vote on the new amendments to the junior doctor contract in 2019.

All the professional bodies have an incumbent advantage when it comes to recruiting members from the professions as they have access to them while they are still students obtaining their qualifications. As long as the professions exist, these bodies will have a conveyor belt of new recruits. This recruitment of young members is contrary to the experiences of most trade unions, with less than one in ten young workers being trade union members. This makes it particularly difficult for the general unions organising in healthcare to recruit members from the different professions, at least in a concerted manner. It is also interesting that the RCN has opened its doors to Healthcare Assistants, a group that UNISON would usually have targeted for membership. Essentially, competition for members is fierce across healthcare work.

The consequence of this is that a substantial amount of its workers are unionised. Membership coverage is enough (approximately 50% of the profession) that each of the professional bodies involved can negotiate on behalf of the entire profession. This can cause conflicts when the interests of two or more professions contradict each other, and may have implications for future pay negotiations.

Our interviewees attitudes to trade unionism varied. One of them was eager to get involved but felt that they were being blocked by their union who did not seem interested. Interestingly, this interviewee was very interested in the Class Composition Project and said that they had tried to do something similar in their previous job. The two youngest interviewees (both bank staff) said that their union had a presence in their workplace but neither of them were members. They both thought that unions were important and had friends in active trade union roles but neither thought that the union could do much for them as they were temporary staff. Another one of interviewees was a member of two unions, GMB and IWGB, and described GMB as something they had essentially as an insurance policy and IWGB as something they believed in. One interviewee had been branch chair at their UNISON branch and said that the branch had gradually grown in strength from a starting point of no officers or reps to one that eventually had a functioning committee and increased membership. However, they were keen to stress that this would not have been possible without a concerted effort from the paid UNISON organisers in the region, perhaps complicating the rank-and-file and bureaucracy split that some union activists see. Equally, the revitalisation of the branch was partly achieved by recruiting members from GMB who had tried to establish themselves (without a facilities agreement) following the initial decline of the UNISON branch. Thus, while the number of UNISON members increased the number of new trade union members did not significantly rise per se.

There are contradictions within union membership in the workplace, as one GMB member explained to us:

I kind of feel like I am a member of GMB, in the same way that I have car insurance. I’m not particularly averse to it. But it’s more to cover, rather than something you have to be a part of. It’s complicated isn’t it? Because sometimes you can have a union which has, you know, a brilliant reputation nationally, but your branch isn’t. And you can have a good branch in a union with a terrible reputation nationally. If you have a really good team at your branch then you can think, ‘Oh, great, yeah, this is really being part of something.’

Taken together, the different attitudes to, and involvement with, workers’ organisations reveal a world where organised labour is present and visible, but many do not feel connected to it. Thus, the question is how to get workers to feel more involved in organised labour; even accounting for general apathy in this regard, there is scope for recruitment. The former branch chair who was interviewed said that they had found crisis (for instance departmental mergers, relocations, and outsourcing) was the best motivator for people to join in the first place, and they could usually be kept on if the union was seen to have done a good job supporting them. This is logical; however, organised labour cannot just be dependent on the advent of crisis to drive recruitment as this does not lend itself to retention and development. While a significant majority of workers in healthcare are covered by collective bargaining (as in times of negotiation the relevant unions and professional bodies are treated as though they represent the entirety of the workforce) at a local level the picture is much more varied and there are many workers who feel no tangible organisation.

However, this does not mean that there is no scope for collective action, and trade union affiliation is not the only form of workplace consciousness. For instance, the two porters interviewed reported that their group of porters (about a third of which were not union members) would regularly take forms of collective action, for instance by refusing to do certain tasks that management would try to add to their roles and, therefore, preventing these new functions becoming part of their job. This would support the idea that there can be an innate form of collectivism brought about by shared experience and interaction in production. This is not an exact science, and there will always be workers who are self-contained, but the potential for collectivism is there, and might be encouraged by how hospital work is often divided into small units in different wards or departments. Indeed, something that could be said about all of our interviewees is that they were all conscious of their class and relation to production if not necessarily all class conscious.

What is the impact of Covid-19?

Unlike many jobs, the bulk of healthcare work did not relocate to the home; however, people’s working environments were still impacted. Many staff were redeployed and had their working environments changed. All of the interviewees said that they thought their employer had responded well to Covid-19 in what were difficult circumstances, and most of the interviewees reported that in the initial weeks of the pandemic there was a lot of uncertainty; for instance, guidance on the correct PPE kept changing. However, eventually, the work settled into a new Covid-19 induced rhythm, production was altered but it was not stopped. Interestingly, they also all agreed that it was nice being able to come to work as it gave them something to do during lockdown and meant they did not have to be inside. One of our interviewees stated that a happy consequence of Covid-19 was that their workspace was moved closer to the hospital’s management, and they were able to build a relationship with one of the senior managers that resulted in positive change. However, the same interviewee also noted that they felt some resentment towards NHS staff from other people (they used the example of taxi drivers) who worked through the pandemic but did not receive the same level of public appreciation or acknowledgement as those working in healthcare. All of the interviewees stated that they found the public support for healthcare workers encouraging at first but eventually it wore thin, particularly when they were denied a pay rise.

Another, albeit temporary, result of Covid-19 for two of the interviewees who were bank porters was that bank staff at their trust were now paid for missing shifts if it was because of Covid-19. Previously, if bank staff cancelled a shift or were unable to attend future shifts because they were ill, they would not get paid anything for their lost future income, now if they said it was because of Covid-19 then they would receive a basic rate of pay.

However, some staff experienced little change in their working environment. One interviewee, who worked in the hospital’s laboratory, said that the only real change for them was wearing a mask, as they already operated in a sterilised environment. Another interviewee started their role during the pandemic and on their first day in the office there was no one else in the office. They did their induction on teams and talked through the layout of the office virtually. This meant that large parts of this interviewee’s jobs were done in isolation.

The advent of Covid-19 also had an emotional toll, and the mental health of workers in Human Health and Social Work activities is noticeably worse than other sectors, with it accounting for 49% of all ill-health in this sector. One of the porters we interviewed, (who was in their early twenties) noted the increased number of deceased patients they had to transport to the mortuary, saying that normally there would be one or two of these patients to move a day, whereas in one of the pandemic’s peaks in December 2020 it was often close to ten a day:

As soon as it hit December, when everything started going belly up again, I was on shift on New Year’s Eve going into New Year’s Day, it was four till midnight. I think on that shift alone, I went up to the intensive care unit about four or five times. It wasn’t old people, there were people in their 30s and 40s. That’s when we knew it was serious, because even during the first wave, it was predominantly elderly people, but then it was people coming through slightly older than you. That was when it started taking a toll on all of us. All we were doing was just going up to intensive care units, taking Covid-19 patients around, or taking bodies down. There was one night me and a colleague took six or seven Covid-19 bodies in a row.

Of further impact to this interviewee was that in the second wave of the pandemic, he was seeing younger patients who still looked older than him (he estimated thirties and forties) but were not ‘old’; they found this deeply upsetting to say the least.

As noted above, one of the consequences of Covid-19 for most staff has been an increased workload as most of the healthcare work was repurposed towards the pandemic and this has contributed to falling staffing levels. Currently, these falling staff levels have yet to manifest themselves in falling output; however, it is unlikely that this is sustainable and likely that this is putting more strain on the remaining staff which could result in more staff leaving in the future. The combination of Covid-19 effectively pausing several NHS services and falling staff numbers, have led to significant backlogs that are still yet to be dealt with and it is not entirely clear when these will be sufficiently addressed.

What challenges or opportunities are on the horizon for organising in the healthcare sector?

One of the most immediate challenges is securing a significant NHS pay rise for staff. A common sentiment across our interviewees was that they felt this was deserved and that the 3% offered by the government was too low; however, they were less sure what an appropriate amount would be. It will be difficult to achieve a significant pay rise. When unions attempted to either ballot for strike action or conduct an indicative ballot, they did not pass the 50% turnout threshold mandated by the 2016 Trade Union Act.¹ If unions cannot strike en masse then it is unlikely that they will be able to secure a significant pay rise for their members; even if strike action were to take place, its outcome would be complicated by the dynamics between trade unions and professional bodies. However, it may be that the current economic climate provokes more militancy at a local level at least (as it is doing across the economy) but the potential achievements of this action, other than a form of catharsis, are currently unclear.

There is scope for successful local action, as is demonstrated by Unite’s recent success at Barts hospital, UNISON’s success at Bradford teaching hospitals in 2019, and UVW’s historic victory at St Mary’s Hospital that secured the in-housing of over 1000 employees. Furthermore, because of the structure of the NHS, successful action at one NHS employer can have ramifications across the whole NHS. For instance, the BMA recently secured a change to maternity policy in a case at a London trust which means that NHS Employers have now updated their advice to all the employers within the NHS. It may be that the most successful way for workers to secure large-scale change would be to conduct disaggregated ballots and focus on large and significant hospitals to best exert pressure on the government.

Outsourcing will continue to be an issue and can be placed in the broader context of workers across Britain having their conditions of employment degraded. The more financial strain the NHS is placed under the greater the incentive for employers to outsource. As with other sectors of the economy, there are signs that the threat of outsourcing (as a moment of crisis) has been a good recruitment tool for trade unions and has resulted in some significant victories, such as UVW’s aforementioned successful campaign at St Mary’s Hospital, as well as St Ormond Street Hospital.

The NHS is highly dependent on precarious labour, all the way from housekeepers to doctors (although precarious doctors tend to have more agency in the labour market due to the nature of their skills). This is a body of labour that unions across Britain have found hard to organise, but this is not to say that this is impossible. Could a union devise a way of appealing to precarious workers and actively facilitating their membership across different employers/branches, then these workers also represent a significant opportunity for recruitment and activism. Disappointingly, statistics for precarious labour were hard to find, whereas statistics for full vs part-time were easy. However, in many respects, the difference between substantive and temporary employment is more important than the distinction between full-time and part-time, as precarious workers have a fundamentally different relationship to production and employment. A worker could be substantively employed for only 50% of the standard full-time equivalent and still have more security (in the form of employment and legal rights) than another worker working more than the average number of hours each week who is on insecure terms and conditions. A potential danger of any data reporting on the quantity of temporary workers in the NHS is that it is not uncommon for NHS workers employed on full time substantive contracts to have additional bank contracts to allow them to do extra hours. This was the case at one of our trusts where Estates and Facilities had their overtime scrapped (as it was paid at a rate of time and a half) and, therefore, had to do all their additional hours on a bank contract that paid the flat rate.

There is also still the spectre of mandatory vaccination. Although the government ultimately changed the legislation (because of the staffing shortages that it would have caused) the threat has not entirely disappeared. NHS employers in England are still expected to gather information about which of their staff are vaccinated and which are not, and there is a chance that in the future, mandatory Covid-19 vaccinations could become a live issue again.

However, for workers across the healthcare sector the most immediate challenge is the increasing workload that has been caused by the pandemic and staff shortages. An increased workload creates more conflict at work and limits agency, and its negative impacts are multiplied with every exodus as the remaining staff take on an increasing workload. While pay increases would be welcome, and are much needed, they will not change the workload and the effects that this is having on workers’ wellbeing, nor will they change the fundamental relationships over production. The point of workers’ organisation is not just to secure a pay rise, it is for workers to collectively finesse the wider processes to which they are subjugated and exert agency over their immediate circumstances.