When is being unemployed not unemployed? A true measure would show more teens are without jobs than people who have supposedly ‘retired’, writes former Treasury senior staffer Tony Burton.

Many New Zealanders feel government is not meeting the needs of the long-term unemployed. Who counts as unemployed remains an issue: A third of your taxes are spent on the welfare system, mostly in payments to support people we believe cannot fully support themselves with paid work, so the definition of unemployment is not an abstract question.

The loosest definition would be the proportion of New Zealanders not in paid work. However this would imply one in three adults are unemployed, and it includes people no society expects to work. Mothers with newborn babies, full time students focused on studying, and 80-year-olds are not going to be filling out job applications.

Therefore the indicator that is popularly used – if a statistic could ever be said to be popular – counts those ready to work and actively searching for a job. In effect it is a measure of the queue for employment. This is the 3.9% unemployment rate announced a few weeks ago, and it’s the right indicator for some purposes such as Treasury predictions of government tax receipts and future government expenditure.

The problem is that the indicator excludes many people who common sense suggests are unemployed: people with disabilities in areas with no suitable local jobs, solo parents struggling to find appropriate childcare, and surfies who live off their parents. In other words, it excludes most of the people we want counted for policy purposes.

An example of just how misleading this indicator can be is shown in the graphs below. The first graph compares the unemployment rate for people in their late teens and those in their late 60s. The teen unemployment rate is increasing over time but is erratic. Little would appear to have changed for people in their late 60s.

The second graph includes the looser definition of unemployment: those who could be working but don’t have a job for a variety of reasons. It shows that far fewer teens are in work than the unemployment rate would have us believe. Meanwhile it indicates that the number of people retiring from work in their late 60s is not 90%, but a little over 55%. People in their late teens are now less likely to be in paid work than those who have just reached “retirement age”.

These kinds of demographic changes raise all sorts of questions, like why does superannuation eligibility starts at 65 if so many people are continuing to work into their retirement years? Whatever the explanations, there is something wrong with the standard measure when it does not even show these profound trends.

This is not the worst of it. The queue for employment is probably the unfairest queue you will ever be unlucky enough to join. Those who have waited the least time are most likely to get jobs. Understandably many in the queue give up, or knowing what the queue is like, don’t even bother joining it. The risk of using the narrow Work and Income “work ready” definition of unemployment is you create a system that’s designed for the people in the queue, because only their outcomes are measured.

So is this an example of the cliché “what gets counted counts”? There is certainly a remarkable consensus that the system is not helping those in greatest need. One of the statements below was written by the National/ACT/Maori Party welfare working group which reported in 2011, and the other comes from the Labour/Green/New Zealand First welfare working group report delivered in May this year. Can you work out which is which?

“There are major deficiencies in New Zealand’s welfare system … This is particularly apparent for some groups, including Māori, young people with few qualifications, disabled people, those who are sick and many sole parents. Addressing these issues requires fundamental change to the welfare system rather than further piecemeal change.”

“We propose [a new] approach based on mutual expectations and responsibilities governing interactions between the state and welfare recipients. It is a commitment to improving wellbeing by supporting positive long-term outcomes for the individual, including increased skills and labour market capability.

Of course, consensus on the problem does not extend to agreement on what to do. I was one of the civil servants assigned to research and draft the 2011 working group report. The language of that report reflects the views of a member whose use of the word “compassion” reminded me of a Dickens character’s claims of being “humble”. Its content is a damning critique of how little the welfare system does to help those in greatest need and provides suggestions for how to improve it.

The more recent report focused on benefit payment rates. However it also made many recommendations on the need to change the welfare organisation and its culture. Its most damning recommendation is simply that MSD “establish an effective employment service”.

For the avoidance of doubt, the first statement comes from the 2011 report, while the second is from 2019.

The unemployment rate is one of a number of indicators that became institutionalised in the first half of the last century (GDP is another one). This means we are using indicators that made sense when the forefront of technology was talkie movies and disability payments were only available to those “of good moral character and sober habits”. The lack of consistent alternatives to these out of date indicators mean the institutions we want to make New Zealand fairer may not be doing that job.

Tony Burton is a former deputy chief economic adviser at Treasury.