25/02/2013 § 2 Comments
And the snapshot makes for grim reading: More than half (54.1 per cent) say New Zealand was a better place to live in 2003 than now. Almost 12 per cent were undecided or could not say.
The article then throws some numbers around, but without much structure or context to give them meaning. For example:
Back then, the average wage was $34,600 – 40.2 per cent less than today’s $48,500.
Averages are less meaningful in this context than medians and some information about income distribution. A look at the Stats NZ Table Builder, though, shows that weekly median income for the working-age population has increased 40% ($401 to $560) while average income has increased only 34% ($539 to $721), which indicates a decrease in income inequality.
These are also nominal dollars, not inflation-adjusted, so the gains aren’t that great. The article tries to cover itself by saying
However, the gain at face value has been quashed, with the cost of living rising considerably over the decade
and then it goes on to quote some price changes for individual products. What we really need is a generalised basket of consumer goods for comparison. Oh wait, Stats already collects that data for us. The CPI is up 26% in that time (2003Q4 = 924; 2012Q4 = 1169). So, the median income gain in that time has been about 11%, or 1% per year.
Of course, these numbers are medians and averages. Polls ask questions of individual people, whose experiences will be across the distribution. Some of them will be worse off, and some will be much better off than 1% per year. Housing is one of those areas:
In 2003, the median house price was five times annual earnings. Now it is almost eight times the average salary.
If you have been owning a house during this period, you are relatively richer. If you haven’t, your relative house-buying power has declined, probably enough to offset any income gains.
My guess, though, is that the poll is less about 2003 than about what happened afterward. Borrowing from Shamubeel Eaqub (NZIER) and a report, ‘Lessons from the Recession’, that he did for MYOB (available here):
We are better off than we were in 2003, but nowhere near where we thought we would be. It’s no wonder that respondents were grim.
24/01/2013 § 5 Comments
A couple of days ago, the Government launched a new website that allows students to compare potential earnings from different university degrees. The Government said it would be a great source of information for students. Student representatives replied that it didn’t tell them anything new.
A word about methodology — this data was made possible by the new types of data and ways of handling it that have been developed over the last few years. By linking actual earnings data from the tax department with tertiary qualifications data, researchers have been able to determine how much students earn after graduation. This data is gold for research on the impacts of tertiary education.
The website allows you to make pairwise comparisons between different levels of study and degree options. Thus, you can find out that earning a Bachelor’s in a foreign language (that was me) makes you twice as likely to be on a benefit a year after graduation than earning an accountancy degree (that was my brother).
In one sense, the students are right. We already knew this. Many studies have shown which degrees earn the most and the least. I’m currently working with a couple of people on a report on returns to tertiary education. It’s been shown that agriculture or humanities degrees have returns between 40% and 90% below average (Psacharopoulos, 2009; Machin & McNally, 2007).
Lucky me, I have both.
On the other hand, the website is a great tool. It makes the information much more available (have you read Psacharopoulos (2009)?) and dynamic and fun. It also makes it more precise. It isn’t just, oh, yeah, those accountancy students will earn more. Students can see that it’s five, ten, fifteen grand a year every year for the rest of their lives.
Will students change what they study? Will they learn something from the new information? That’s the great thing about this data — we will be able to find out. In five years’ time, we can look back and see whether the composition of degrees in New Zealand has changed.
Also, that research will tell us something else: how much study is preparation for the workforce versus self-improvement or simply consumption. The more students see study as a path to higher earnings, the more the new information should change behaviour. Maybe it won’t change behaviour after all; maybe students already know this stuff; maybe students are already doing what they think best.
28/05/2012 § 1 Comment
The Green Party responded to the Budget with this graphic:
I carry around in my head a vague sense of the income distribution of New Zealand. The Green’s figures did not match that distribution. So, I went looking for the numbers. Helpfully, the Party’s website had more information (pdf) about how these incomes were calculated.
- Solo parent: ‘$333 a week DPB, student allowance and housing allowance of $125 per week, $80 working income, and $88 Family tax credit. This actual puts her in the middle of the 3nd decile – ie 20% of household [sic] have lower incomes.’ I can’t verify the income because I’m not familiar with these government programmes. The additional comment, that 20% of households have lower incomes, is misleading. According to the NZ Income Survey, 15% of households have one member so may not be poorer per person. Also, some of these payments are specifically aimed at lifting children out of the poorest deciles, which they appear to be doing.
- The typical family: $71,500 seems high for the ‘typical family’. The Greens link to the NZ Income Survey, which reveals that the figure is indeed the median for a family of at least five (‘Couple with three or more dependent children’). They aren’t really ‘typical’ — they are only 5.7% of all households, and only 14.0% of households with dependent children (there are more solo parent households — 16.4% of households with dependent children).
- The top 10% family: The additional information states that the $250,000 was ‘Estimated from Household Income Survey (Income) data’. I don’t know how they ‘estimated’ this figure, but it is a fairly high estimate of the top 10%. IRD income statistics make this clear. Only 1.4% of earners (in 2010) have an income of $150,000 or higher. Assuming that the part-time earner can only make $50,000, then the main earner needs to bring in $200,000; there are 0.7% such earners. So, this isn’t a top 10% family but a top 1%.
- The top 1%: This is the average taxable income per earner for the top 1% of earners, using IRD figures. However, it isn’t a household figure (like the others are) and it isn’t a median (which would be a better representation of this distribution).
The IRD figures don’t tell us about families and households because they are only taxable incomes and include earners like students who are working part-time. But, they are a guide. More than half of earners make less than $25,000. The top 10% of earners make about $70,500 or more — the top 10% of ‘families’ will be higher.
What all this means is that the Green Party figures are skewed upwards. What they are presenting as ‘typical’ isn’t. Even when they talk about the ‘top 10%’, they overestimate the income. So, why?
I can make a few guesses:
- It probably maximises the increases and decreases they calculate for the various families. So, the choices are political rather than economic.
- They may be trying to emphasise the distance between the majority and the top 10%. However, they are simply wrong. People making over $100,000 are under 5% of earners, so there are still lots of the top 10% making under that.
- They don’t know how lucky they are. Many of the people who brought you this analysis are themselves in the top 10%. But life is expensive — housing, kids, cars, insurance, etc. They don’t feel rich. They can’t quite believe that 90% of the country toughs it out on less — a lot less. So, they recalibrate their mental model of the income distribution with themselves in the upper middle instead of the top.