4.2 Total Cost Elements
4.2.2 Human Capital Costs (HKC)
Conceptual and methodological issues
One of the most significant costs of workplace injury and disease is the long-term loss of work and productivity. There are two broad approaches for measuring this loss described in the literature:
- lost wages/human capital method
- friction method.
The lost wages/human capital method is based on neoclassical economic theory. Wages and other marginal costs are assumed to equal the value of the marginal revenue generated by an additional worker under conditions of full employment.22 Lost product is thus the value of the wages (measured as average earnings) plus other inputs to production (capital, plant and equipment, land, enterprise etc) multiplied by the number of work days missed. For reduced productivity while working, a percentage of this calculation is used.
The limitations with this approach are that it does not incorporate quality of life dimensions, and the choice of wage rate is an issue. Most studies use an all-industry average wage rate, whereas some use a minimum wage or industry-specific rate. The choice of time period is also an issue, with some studies using a fixed period and others using a variable period (such as life expectancy). It also does not allow for the situation where a previously unemployed person takes on the work of the injured employee. In this situation, society may not suffer as large a loss, as the previously unemployed worker who generated no income now generates an income, while the injured worker no longer generates an income. This leads us to the alternative friction method of calculating productivity losses.
The friction method was developed by Koopmanschap, Rutten, van Ineveld and van Roijen.52 This approach estimates production losses for the time period required to restore production to its pre-incident state. This contrasts to the human capital method where, for example, the potential loss of production from a disease is calculated over a much longer timeframe. The time period used under the friction method could be determined by when the employee returns to work, or by when a replacement is found.21 This method generally assumes that there is unemployment, and that a person who was previously not earning an income replaces the injured worker.
Predominantly the human capital/lost wages method is used, and indeed this is what Leigh et al9 used in the US and NOHSC/Industry Commission in Australia. The reason for choosing the human capital method tends to be a general recognition that, after the initial disruption, until production is restored to former levels (most relevant for minor injuries), there is essentially the loss of the labour resource (when there is permanent disability or fatality) over the longer term, which reduces the capacity of the economy to produce at any given level of unemployment.
This human capital approach is also considered appropriate in New Zealand, as a developed economy operating at near full capacity, and that the human capital losses are not sufficient to influence the average wage.
For employees who only return to work on a reduced basis, there is the issue of their earnings profile. Weil19 depicts three earnings profiles:
- Minimal economic loss – where the employee re-enters the workplace with a lower level of earnings (prior to their accident), but where their earnings then rise quickly to the pre-injury earnings profile.
- Moderate economic loss – where upon re-entry to the workforce, at a lower level of pay, the post-injury earnings profile has a reduced rate of earnings to that which would have otherwise ensued.
- Significant economic loss – where there are multiple entries and exits from the workforce and the absolute level of earnings falls each time.
According to Weil, a number of studies have looked into earning profiles associated with injuries and return to work schedules. In particular, Weil cites Reville23 and Peterson et al53 to state that, over the five years following an injury, workers receive approximately 40% lower earnings on average than an equivalent control group of employees.
Section 2.3.4 summarised the approach to distinguishing human capital and productivity measures from the more recent and comprehensive literature on willingness-to-pay methods that enable valuation of the non-financial aspects of the value of leisure, health and human life.
NOHSC18 correctly considers the human capital cost (HKC) as the stream of lifetime earnings that is lost when workers with an injury or illness are unable to re-enter the workforce at their former level, until average retirement age. This is the difference between the present value of the earnings before the incident – PV(EBI) – and the present value of earnings after the incident – PV(EAI). Mathematically:
The sum is over the period of years remaining – that is, retirement age (RETA = 62 years in New Zealand) minus the average age following the incident at which the worker returns to work (RTWA = 40 years in New Zealand).
- The RETA parameter estimate (which is in fact the same as in Australia) is drawn from OECD.vii, 54
- RTWA is derived from the ACC data (it is three years younger than in Australia) based on the average age of an incident (39.6 years) plus the average time off work (175.1 days).
The summations are thus over a period of 62 – 40 = 22 years. The discount rate is 3.8% as derived in Section 2.3.4.
A few points are worth noting:
- For all workers, EBI = AWE*52 (where AWE is the Statistics New Zealand estimate plus the 12.5% on-cost loading).
- For category 3, 4 and 7 workers, we use the NOHSC18 estimate that EAI(3,4,7) = 0.64 * EBI for workers experiencing a partial return to work. This may be conservative since some of the “other” workers may not, in fact, return to work.
- For category 5 and 6 workers, EAI(5–6) = 0 since there is no contribution to production from those workers after the incident.
- The sum over all injured workers will depend on the distribution of their ages on leaving the workforce, as well as on their average age. This is because the net present value of lost future earnings varies non-linearly with age. However, we do not make this refinement in this analysis.
- $46,498 is the average pre-incident wages, so the present value of the future incomes stream is $711,001.
- $29,758 is the average post-incident wage (64% of the pre-incident wage), so the present value of the future incomes stream is $455,041 and the lost earnings are $255,960.
HKC by bearer of cost
None of the HKC is borne by the employer. It is borne in part by the worker, as loss of income, and in part by society, through compensation and welfare payments.
As with PDC, it makes sense to calculate the worker’s loss of earnings as a residual – the difference between the total HKC and the welfare, compensation and tax losses borne by society (since the employer burden is zero).
HKC borne by society
Average compensation payments for longer-term loss of earnings for severity categories 3 to 7 (a matrix we call CPWL(3–7,c)) is derivable from the ACC data. As with the CPWS matrix for PDC, this matrix is zero for uncompensated workers. The “long term” is defined as the period after the workers are permanently replaced (partially or fully).
As noted earlier, the provision of welfare and tax payments is complex, detailed in 4.2.5. We call the long-term welfare payment matrix WPL and the long-term tax matrix TAXL. So:
Total human capital cost is thus estimated as $3.05 billion (Table 4.7) of which:
- $1.88 billion (62%) is borne by the worker
- $1.17 billion (38%) is borne by society.
| TABLE 4.7 | HKC by severity and bearer of cost, 2004–05 ($m) | ||||
| SEVERITY | TOTAL | EMPLOYER | WORKER | SOCIETY | |
| <7 days | – | – | – | – | |
| Full return | – | – | – | – | |
| Staged return | 1,911.7 | – | 1,109.0 | 802.8 | |
| Partial return | 66.2 | – | 38.0 | 28.2 | |
| Permanent | 26.5 | – | 17.6 | 9.0 | |
| Fatal | 675.5 | – | 522.4 | 153.0 | |
| Other | 369.7 | – | 194.5 | 175.1 | |
| Total | 3,049.6 | – | 1,881.5 | 1,168.1 | |