Most of what people think they know about income inequality is wrong, writes Eric Crampton of the NZ Initiative.
Rather than rising since the 1980s, income inequality in New Zealand rose in the late 1980s and early 1990s, then flattened out. Some of that increase was real, but some of it was complicated – as explained late last year.
Pointing out the flat income inequality statistics typically invites objection that wealth inequality has worsened. But there too, a lot of what people think they know about it is wrong.
Unfortunately, data on wealth is far worse than data on income – the Government gathers a lot less data on wealth. For example, Statistics New Zealand reports that people in the least wealthy 20 percent have $1.75 in property debt for every $1 in property assets – but no bank in the country would extend a loan on that basis. It is more likely that the survey data misses some houses owned through family trusts where, for example, a 25-year-old takes over the mortgage and effective ownership of their parents’ second home held in the family trust. The mortgage payments and mortgage debt are noted in Household Economic Survey data, but the ownership may be missed. This means their parents’ net wealth will be overstated, their own net wealth will be understated, and measured wealth inequality among younger cohorts would be somewhat understated but overall measured wealth inequality would be somewhat overstated.
And the discussion is not helped when organisations like Oxfam release reports on wealth inequality that could provide case studies for university courses on how to lie with statistics. But we’ll come back to that.
Let’s work through what we know, starting with basic principles before we get into the numbers. Even the basic principles quickly lead to counterintuitive results.
When we talked about income inequality last year, I noted how incomes change over the life cycle. People typically start on low or no wages while they are in training, earn more as they gain experience, save during their peak earning years, and spend the proceeds of those savings while on lower income during retirement.
That process automatically builds a life cycle around wealth.
A lot of people will start out with student loan debt and will report negative net worth because of it – about 5 percent of households in New Zealand then report negative net worth. It’s a bit strange. If you borrow money to buy a business that provides earnings of $50,000 per year, the capital value of that business counts against your loan when figuring out your net wealth; if you borrow money to invest in a degree that provides the same increase in earnings, you’re only counted as having debt. This kind of thing will matter a lot when we look at work like Oxfam’s.
Over the course of their careers, people typically pay off their student loan debt, pay off their mortgages, build up a retirement portfolio, and then spend down that portfolio. It isn’t true for everyone – some will start with inherited wealth, and others will never build up a retirement portfolio. Luck can matter too along the way. But the basic life-cycle pattern matters when looking at the overall data.
All of that means that, if you took a hundred people who had identical career paths and identical inflation-adjusted earnings along the way, but who were of different ages, you would have substantial measured inequality in wealth – but that inequality that would be of little moral consequence.
Good measures of wealth inequality should attempt to adjust for this life-cycle phenomenon – especially if demographics change over time or if you are comparing groups that have different age profiles. The Tax Working Group noted that wealth inequality is highest among 25 to 34 year-olds, lowest among those over the age of 65, and about half as large as wealth inequality across the population as a whole within older cohorts. Age patterns matter.
It sounds obvious that we should correct for age patterns, but not doing so is a rather easy mistake to make. For example, 2017 work by Max Rashbrooke pointed to substantial differences in wealth between Kiwis of European descent and Kiwis of Maori and Pasifika descent, but ignored that the median age for Europeans in New Zealand was 41 years (based on the 2013 census), while the median ages for Maori and Pasifika were about 24 and 22 years, respectively.
Comparing ethnic groups without adjusting for age differences will lump together the effects of ethnicity with the effects of being younger. Most people have rather more wealth at age 41 than they did at age 22 or 24. The Tax Working group showed that ethnic differences in wealth remain after adjusting for age, but the gaps are smaller.
Good measures of wealth inequality should also try to account for relevant changes in policy – or differences in policy when making comparisons across countries. Perhaps surprisingly, policies like government superannuation schemes can mean that measured wealth inequality is higher than it otherwise would be – even if retirees are just as well off under private savings accounts. Why? Private savings accounts count towards measured wealth but superannuation entitlements do not. And the same is true if you compare countries with public health systems with countries with medical savings accounts.
Because New Zealand has a comprehensive superannuation scheme, it makes little sense for workers in the lowest quartile of earnings to save much for retirement – payments through superannuation are not that much less than their earnings while in employment. Measured wealth inequality is consequently higher than it otherwise would be; workers’ implicit savings through the Government’s superannuation scheme do not count towards their wealth. The wealth-equivalent of that superannuation entitlement is not small: someone aged 65 and expecting to live for another 15 years has a superannuation entitlement worth about $225,000 – a pretty hefty figure when median net worth for those aged 65–74 was $416,000 in the 2018 data.
Unfortunately, it is impossible to find internationally comparable measures of wealth inequality that even attempt to work in these kinds of entitlements. So international comparisons should be viewed with some caution if there are substantial differences across policies like superannuation.
Statistics New Zealand’s 2018 household balance sheets show that the wealthiest 1 percent of households own 16 percent of overall net wealth in New Zealand (down from 20 percent in 2015); the top 5 percent own 39 percent (down from 40 percent); the top 10 percent own 53 percent (unchanged); and the top 50 percent own 94 percent of net wealth (up from 93 percent). There has been little recent change in wealth concentration.
But while their headline figure contrasting the increase in net worth among the richest 20 percent with stagnant figures for the bottom 40 percent make it sound like rich people are scooping up all the wealth, the finer-grained figures instead show more generalised increases in wealth. Median household wealth increased from $289,000 to $340,000. In 2015, 8.3 percent of Kiwis households had more than $1.5 million in net wealth; in 2018, 12.5 percent of Kiwi households were in that group. Fewer households held between $100,000 and $500,000 in wealth and more households held assets worth over $500,000. About 30 percent of households held less than $100,000 in net wealth in both years, but remember too that people move between categories over time: every year brings a new crop of university graduates with a lot of debt and few assets, and a new cohort of workers who have paid off their loans and are moving into the higher wealth brackets.
The Tax Working Group also included this graph putting New Zealand’s wealth concentration into international context:
The wealthiest 10 percent in New Zealand have a smaller proportion of overall measured wealth than their counterparts in Denmark, the Netherlands, or the United States; have about as great a share of overall wealth as the wealthiest 10 percent in Canada, France, the UK, Ireland and Austria; and a greater share than in places like Japan and Italy. Our measured wealth concentration is higher than that in Australia – but you will remember my note earlier about countries with a government-run superannuation scheme and countries that rely more heavily on individual savings schemes.
So when we look at Statistics New Zealand’s data, wealth inequality is little changed over the past few years, and wealth concentration in New Zealand is about on par with the kinds of places we’d expect to be matching.
And all of that brings us to Oxfam’s report released last week. Oxfam argued that wealth inequality in New Zealand is increasing and pointed to the disparity in the growth in wealth enjoyed by the two Forbes billionaires who live in New Zealand, and the drop in wealth for New Zealand’s bottom half.
Oxfam’s wealth inequality reports have a lot of problems every year – but don’t just take my word for it. Tim Harford’s 2016 critique of the Oxfam reports on BBC’s More or Less still holds – as does Noah Smith’s more recent critique from the centre-left.
It is almost as though people want to believe that the rich are getting richer and that the poor are getting poorer, and that it’s worse here than elsewhere, regardless of the data.
But a few things should strike you immediately about this year’s New Zealand report. Statistics New Zealand shows that median wealth is increasing. That means the wealth of the person at the top of the bottom half has increased. But Oxfam argued that wealth of the bottom half had decreased substantially. What happened? And why did none of the reported increases in wealth at the top show up in the reported wealth of New Zealand’s top percentiles?
Oxfam’s reports rely on Credit Suisse’s annual compilation of global wealth statistics. Credit Suisse attempts to make international comparisons by putting all wealth holdings, including those of Forbes billionaires, into their US-dollar equivalents. New Zealand experienced a drop in the exchange rate that reduced the US-dollar value of all New Zealand wealth denominated in New Zealand dollars. So according to the Credit Suisse figures, both the bottom and the top halves became less wealthy over the past year. Oxfam’s playing this as a drop that affected only the bottom half may have been an oversight, a bit cheeky, or so deeply misleading as to count as lying – depending on your point of view.
More interestingly, Credit Suisse reported a lower overall Gini estimate of wealth inequality for New Zealand in its most recent report than it did in the prior year’s – a drop from 72.3 to 70.8. The drop certainly was not large, especially considering the effects of shifting portfolios into US-dollar equivalents and methodological changes along the way, but it was markedly different from the pan-media headlines advertising rising wealth inequality.
While the Tax Working Group’s initial assessment was far less credulous than the recent media commentary, public misperceptions about the true state of the world drive demand for poor policy. It is almost as though people want to believe that the rich are getting richer and that the poor are getting poorer, and that it’s worse here than elsewhere, regardless of the data.
The Credit Suisse reports do help provide some perspective.
Half of all adults in New Zealand can count themselves among the world’s top 10 percent, and 189,000 of us are among the global top 1 percent. And if we can ever fix our broken housing markets, we’ll see greater improvements at the lower end of the distribution.