Demand and Overwork – Employee Sickness/Absence as Resistance
Flight crew salaries make up between 5.9% and 10.5% of airline direct operating costs (IATA, 2017). In my hierarchical model, the current gross availability of pilots is influenced by environmental factors. Forces such as the reduction in the scale of military aviation (and, hence, the supply of trained pilots), increasing costs of pilot training, reduced access to funding from banks for training after the financial crash of 2008 and the rapid expansion of commercial aviation in Asia (with the subsequent increase in demand for pilots) are all beyond the control of individual airlines. What airlines can control is the utilisation of crew: exploitation of the asset.
The expectation of effort imposed by management on the workforce is reflected in the design of the work schedules that sometimes rely on overtime, duty period durations and constraining time off or controlling access to contractual vacation allowances. We saw earlier that recruitment and training can lag behind the operational requirement thus creating a potential for overwork. Not advertising for a replacement until the present incumbent has left can also create a temporary vacancy, thereby representing a saving on staff costs. In aviation, flight time limitations represent a legal constraint on pilot utilisation and, thus, available production capacity. Defining crew productivity requirements represents an archetypical example of ‘control’ as framed earlier. While the contractual obligation may be defined in units of time, the consequences of how that contract is met place a significant burden on worker health and well-being. Where contractual demands are perceived to be either excessive or unreasonable, such that employees feel overworked, the classic response is absenteeism.
Company policies on eligibility for paid sick leave and absence management derive from state regulations, which differ considerably between the states. Companies can choose to offer more generous paid sick leave and more flexible absence policies than the required minimum, but the interplay between the regulatory frameworks and individual behaviour is actually quite complex. Establishing policies to deal with both sickness and absenteeism is a legitimate management function.
There are a variety of reasons why the crew absent themselves from the workplace. One is, of course, genuine ill health or a significant medical condition and is what we could call involuntary absence. The other main category of absence is voluntary: the employee chooses to not report for duty. Voluntary absence may be legal, an example being allocated vacation, or it may be illegal. An example of an illegal voluntary absence w'ould be a decision to prioritise personal requirements, such as a family commitment, over the obligation owed under the conditions of employment. Another cause of illegal voluntary absence is disaffection. For whatever reason, the employee holds the employer in such low regard that absence is a form of private industrial action, not to be confused with more organised attempts to disrupt operations by mass ‘sickness’ events, for example.
Unlike the broader working population, there are several reasons why a pilot might be unfit while another worker might still be available, a head cold is one example. Pilots are also required to hold a valid medical certificate, and various rules relating to the issuing of a certificate also render the pilot unavailable for duty (minimum recovery times for specific conditions, restrictions on who can certify fitness). Pilots who are off work for more than a certain period are required to seek recertification. As a consequence, pilots might be fit for duty but still unavailable. Such pilots are deemed long-term sick. It is difficult, if not impossible, to make direct comparisons of lost productivity between the airline pilots and the general working population. Factors such as state rules on access to paid sick leave and the general job market affect absence rates.
Attendance represents a contested space between the workers and management where tensions over demand are visible. My first encounter with this issue was w'hen I looked at daily operations in a UK low-cost carrier in its first year in business. Relations between the management and the 99 cabin crew employed by the airline were poor but the company did not recognise unions, and so the cabin crew felt that they had little representation. In response to perceived increased rates of absenteeism, management issued instructions that annual leave would be awarded on a point system based on attendance records. The new policy came into effect at the end of the summer season, just as the weather started to change and a common time for colds and flu. The crew' now started reporting for duty with head colds, and several had issues w'ith clearing their ears in flight, with their balance, and some were so ill that the captains refused to let them fly on the return sector. Over a 2-week period, nine cabin crew were offloaded at destinations, causing serious disruption, including passenger load restriction on return flights because of cabin crew' to passenger ratio limitations and flight re-routing to collect stranded passengers. Another airline I worked w'ith actually issued a policy that said ‘colds and flu was not a sufficient reason for failing to report for duty’ while another stated that ‘stress’ was not an acceptable cause for absence, just as the head of cabin crew departed for 6 months of stress-related sick leave.
The interplay between policy and behaviour can be quite complex. Figure 9.1 show's pilot absence rates and pilots with unbroken annual attendance records (not absent) over a 10-year period for a medium-sized airline. Both sets of data suggest a change over time w'ith absence rate increasing as more pilots took days off. While interesting, the data do not reveal the cause.
FIGURE 9.1 Pilot absence.
In order to understand the data, we need to look at the distribution of absence events. Patterned absence is a common feature of employee behaviour. For example, people are often ill on a Friday or a Monday. Bank holidays often result in spikes in additional time off. These patterns are common across all areas of employment. Figure 9.2 shows the distribution of absence events per pilot for the same airline over an 11-year period. The number of pilots having no absences declined over the full range of the data set. (Figure 9.1 shows this as a rate for all pilots while Figure 9.2 shows the rate as a percentage of all absence episodes). The rate for one or two episodes/pilot remained fairly constant, which is to be expected given that humans do fall ill. There is a change in the number of crew taking three or four episodes and again for five or six episodes a year with the trend generally upwards for each category.
Figure 9.3 shows the distribution of episodes by duration. Again, the data show that taking 1 or 2 days off at a time remained fairly constant while absences of 4 days duration show a gradual decline. The episodes of 3-day duration show a remarkable
FIGURE 9.2 Absence events per pilot.
increase in 2016, coinciding with a decline in the rate for episodes of 6-day duration. While the pattern for events shows an increase over time, the data for episode durations shows a change. Broadly speaking, a couple of days off a couple of times a year might be considered normal wastage, but crew appear to then taking more episodes of 3- or 4-day duration.
For the period from 2008 to 2015, the absence management rules in force at the airline were six events or 30days total in a rolling 12-month period. The rate for six episodes in Figure 9.2 probably reflects the difference between a calendar year and a ‘rolling 12-month’ period. It is possible to be absent every 2 months and not trigger the threshold. This is possibly the first clue that pilots are able to game the system. In 2016, the rules were tightened up to try to reduce the rate of absence. Trigger events were set at five episodes in 12 months and a total of 21 days. In addition, whereas 7 days of self-certified absence per episode was permitted under the old rules, this was changed to 3 days, after which a doctor’s certificate was required. This change in self-certification policy seems to be reflected in Figure 9.3. Absence durations reflect the rules for submission of a sickness certificate. To explore the impact of this policy change in more detail, Figure 9.4 shows the data for a specific fleet for the year in which the self-certification limits changed. The new' policy w'as introduced in April. There is a clear shift away from episodes of 7-day duration to the new 3-day limit.
Absence is a legitimate concern for management. No-notice absence is disruptive and unwarranted absence represents moral hazard: a theft of time and wages. Looking at episodes and durations together, patterns could be detected of which the most common w'as the ‘4 and 20’ rule: 5 days off every 3 months. The fact that there were clearly discernible patterns of absence across the year can only suggest intentional behaviour. The willingness to fulfil contractual obligations is a barometer of the state of the relationship between the management and workforce. It could be seen as an indicator of morale. The data suggest that pilots respond to management interventions, but they still do not address the cause. In these patterns, we see pilots coping with genuine illness but also, in all probability, some are using absence to recuperate from the demands of the task or manage the competing demands of family life. Others may be expressing their level of disaffection w'ith the organisation or choosing to place self-interest ahead of contractual obligations.
FIGURE 9.4 Number of absence episodes (% of all episodes).
Policies and procedures are tools that can be used by management to exercise control, but this example suggests that rather than control an issue, policies shape the nature of the response. There is a tension between the management of legitimate action and abuse. In 2018, the UK confidential human factors reporting programme (CHIRP) received several reports from the crew who felt that absence management rules were being used to apply undue pressure in order to coerce crew to operate when unfit. At the time of writing, a study is underway to consider the implications of absence policies from a safety perspective.
As we saw in Chapter 4, aviation makes significant demands on the individual, and ill health is often the result of succumbing to those demands. Absence from the workplace is usually legitimate but, where it is not, it is often unavoidable. The issue we have been considering here is that the space between the management and workforce where individuals exploit rules out of personal interest but also as an expression of disaffection. The next area of tension between the management and pilots is what Hodson called ‘challenges to autonomy’.