Analysis by Keith Rankin.
Before the 1950s
My MA thesis was about labour supply in New Zealand and Australia during the 1920s and 1930s, with particular reference to the 1930s’ ‘Great Depression’.
The central finding related to the workings of whanau as an economic unit, and how the loss of income to the ‘family breadwinner’ – in those times, almost always an adult male – created a change of labour force participation status amongst the other adults and teenagers in the family. (Note that the word ‘whanau’ essentially means ‘family’. Here, it means family as an economic unit; so, in a Māori context it may mean ‘extended family’, whereas in a Pākehā context it more likely means a ‘nuclear family’.) This propulsion of erstwhile dependents into the workforce is known as the added-worker effect.
In the 1920s, modern labour force concepts started to take hold. It became clearer than ever before who was in the labour force, and who was not. The 1920s represented the end of a half-century in which, increasingly, economic progress was measured by how many dependents a worker (‘breadwinner’) could support. Thus, socio-economic status was conferred on families with large numbers of people who were clearly ‘not in the labour force’; we may call these people private beneficiaries. In the 1920s, it was normal for female school leavers – especially ‘respectable’ lower-middle-class ‘girls’ – to return home, and stay there until they married. (Today they would be called NEETs; in relation to productivity, NEETs are now seen as ‘the problem’ rather than as ‘the solution’.)
This all changed in the 1930s, with the advent of the Great Depression, although it continued to be very difficult for married women to be acknowledged as anything other than ‘dependents’; indeed strong social barriers prevented married women from being wage or salary earners.
The ‘added-worker effect’ is what happens when a loss of family income (or a rise in household costs, such as rents or mortgage interest rates) – causes family dependents to enter the labour force as support breadwinners, as ‘additional workers’. It means that the actual ‘labour force participation rate’ increases, even if the reported participation rate does not increase. For example, a woman who had been happily dependent while her husband was employed, becomes an unemployed person after her husband loses his job; she becomes available for work, including the precarious self-employment we most associate today with developing countries. Such recategorization is officially recognised under ILO (International Labour Organisation) definitions; for example, in New Zealand today such a couple would both qualify for the ‘Job-Seeker Benefit’ as unemployed persons. However, in the pre-feminist era of sex-defined economic roles, her ‘unemployment’ was not statistically recognised as such.
In the Great Depression, there was a very large entry into the labour force of married and single women. Much of this was hidden, because many of these women were unacknowledged unemployed. There was also an increase in the teenage male labour force; many teenagers of both sexes quit school so that they could contribute to the family finances. (Textile factories ran hot, employing cheap teenage labour.) In many cases, this workforce participation would involve the teenagers leaving home to become remittance workers. The labour force also expanded through many older male breadwinners staying in work; for financial reasons they were not able to retire, as they otherwise would have done.
Before discussing more recent times, it is important to note that the added-worker effect has a converse, which we may call the subtracted worker effect. Indeed, it was the productivity growth and increased prosperity from the 1850s – in the United Kingdom in particular – that enabled ‘decent’ (working class) girls and boys to withdraw from lives of remunerated toil, and to attend school; and women withdrew from income-generating activities in favour of home-based activities. It was also at this time that the concept of ‘retirement’ became established.
Further, it was in these years – from the 1850s to the 1920s – that there were progressive reductions in the numbers of lifetime hours a male breadwinner would be required to commit to income-generating activities. In New Zealand in the 1940s, we saw a truly dramatic increase in the number of men showing up in the census as ‘retired’. This was both due to the rapid economic growth that took place after 1934, and the introduction of Universal Superannuation from 1940.
From the 1950s until 2020
In the 1950s and 1960s, New Zealand’s women were much more connected to paid work than the official statistics suggest. The Great Depression created a new mindset; indeed in many respects feminism as we would understand today emerged in the 1930s. Women who had been employed in the 1930s no longer expected to live as their mothers did; they expected to return to paid work once their future children were no longer fully dependent on them. By and large, the mothers of the ‘boomer’ generation participated actively in the labour force both before and after their lives as full-time child-raisers. (The ‘before’ was often very short; women born in the 1930s and 1940s tended to start their families very young by today’s standards.)
Since the expansion of the post-war welfare state, the ‘added-worker’ dynamic has changed; for some people, the alternative has been to contribute to the household economy through the benefit system. (And married women would not want to disqualify their sick or unemployed partners from receiving benefits.) Increasingly, in more recent years, to receive a benefit a person and their partner must both be in the labour force; ‘married’ couples – for the most part – can only receive a public benefit if they are both ‘job-seekers’; so the dynamic of one partner seeking income when the other becomes redundant is as strong as ever.
While – in post-war times – the added-worker effect remains as relevant as ever, the subtracted-worker converse effect has substantially diminished in the face of a labourist culture that is a legacy of the Depression. As a result, before the onset of Covid19, just about everyone over 15 was either in employment, unemployed, or preparing for employment. (Secondary education these days is seen as ’employment preparation’.) Further, in the 2010s, a higher percentage of people aged 65 to 74 were earning labour income than in any other post-war decade.
In the post-war years – and especially from the 1980s – there has been a ‘ratchet effect’ whereby each period of high unemployment has added to the labour force participation rate, while the times of low unemployment have seen minimal change to labour force participation. Further, we have come to believe that what we have been getting is desirable; we now believe that high and growing labour force participation is an economic goal to be achieved, and not (as in the later nineteenth century) a problem to be solved. In the Policy Targets Agreement – the contractual arrangement between the Government and the Reserve Bank – the Bank is required to conduct monetary policy with a view to ‘maximising employment’, which is distinctly different from ‘minimising unemployment’.
In the 1980s it was the high mortgage interest rates that made two-income families the new normal; and that new normal meant that renters could pay more because renting couples both had an income. (That new normal meant couples would delay having children, and mothers would use childcare for their preschool children.) The subsequent unemployment and welfare benefit cuts – especially in the early 1990s – drew more additional workers into the labour force; this time, as in the Depression, many of the new workforce entrants were hidden – they were unemployed but not included in the unemployment count.
Nevertheless, despite historically high pre-pandemic participation rates, the added-worker effect is back. What is happening in Pasifika communities in Auckland is very reminiscent of the Great Depression. (Refer Covid poverty for hundreds of Auckland’s Pasifika Radio New Zealand, 3 September.) Pasifika – much more than Pākehā – run household economies not unlike those in New Zealand in the 1930s. Indeed, these communities have always had remittance obligations. Now Pasifika teenagers are leaving school in order to help their parents to meet their financial obligations.
The added-worker effect should be statistically visible from the Household Labour Force Survey, showing up as people transiting from ‘not in the labour force’ to either ’employed’ or ‘unemployed’. However, there is a conceptual anomaly which has been built into the definition of ‘not in the labour force’. This is the anomaly of the ‘discouraged worker’; the anomaly that actually led to a technical fall in the unemployment rate for the June 2020 quarter.
The ‘not in the labour force’ statistical category is meant to be a measure of people in ‘retirement’, ‘caregiving’, ‘fulltime education’ and ‘voluntary work’. The statistic is flawed for two reasons. The first reason is that many people may be a mix of these activities, and any evidence of paid employment trumps all of these. Thus retired workers in higher education who are looking after grandchildren and doing voluntary work will be counted in none of these activities if they are doing as much as one hour of paid work, or doing unpaid work in a family business. Thus, the measure has a bias towards counting people as ’employed’. The problem here is that people who belong in multiple statistical boxes – that is, most people in the real world – are being categorised on the basis that they may only be attached to one box.
The second statistical flaw is that unemployed workers who are deemed to be not looking hard enough for paid work, or who are not able to start a new job ‘immediately’, are discarded from the unemployment statistics. These people are called ‘discouraged workers’; they are regarded as ‘voluntarily’ jobless. Their presence means that official unemployment statistics are biased downwards, making unemployment seem to be less than it really is. Many NEETs – ‘not in employment, education or training’ – are, statistically, discouraged workers.
Perhaps of even more concern than the unemployment undercount, discouraged workers (a deduction from the official labour force measure) become more numerous at precisely the same times that additional workers (an increase to the official labour force measure) also become more numerous. Thus, these two important groups cancel each other out in the labour market statistics, rending both invisible.
Statistics are used to inform policy. Flawed concepts and measures inform bad policy. We have no shortage of unimaginative and poorly informed economic policy, from all sides of the political spectrum.
The statistics that I first produced in my MA thesis have been used by historians to indicate the scale of ‘unemployment’ in the Great Depression. They are usually cited with qualification, as applying a ‘generous definition’ of unemployment. In my thesis, I did not call them ‘unemployment’ tallies. Rather, I used the nuanced term, ‘residual labour force participation’. In fact, most of the unemployed during the Great Depression would not qualify to be counted as unemployed, if a strict application of modern statistical definitions were to be applied.
The problem is not that unemployment was less of a problem in the Depression than we previously thought. Rather, unemployment in the Depression was a substantially greater problem than either contemporary statistics or modern definitions would have us believe. People did not spend their time applying for jobs, as we understand ‘jobs’ today. Rather they hustled for income; everyone in the many affected families, not just the appointed breadwinner. The trick to understanding the Depression was to see what people were doing, not what they were not doing.
It’s becoming the same today, and not just because of the Covid19 pandemic. The labour market is becoming much more precarious, ‘casual’, fragmented. And many more people than before are in some respects self-employed. Post war labour market statistics – and academic analyses – focus far too much on wage and salary workers – and far too little on the precarious self-employed. So much of what happened in the Great Depression would have been counted today as self-employment rather than as unemployment; it is becoming so again. Casual self-employment – then and now – was often what the unemployed tried to do, because they had no other option.