Thursday, October 29, 2015

Two-child China, and population ageing myths

China abandoned its one-child policy yesterday. Just about everything I’ve read since explains that this policy shift is a result of fears about dependency ratios; the ratio of the number of non-working age people in the population (children and the elderly) to the number of working age people. As shown in the chart below, China, like most countries, is seeing the start of an uptick in this ratio due to an ageing population.

But the simple fact is that increasing fertility rates isn’t a solution to this problem.

The reason population growth doesn’t solve this problem is that a growing population relies on
  1. more children, and hence a higher youth dependency ratio, or
  2. more immigrants, who become elderly themselves, delaying the problem. 
The only way population growth can ‘solve’ the age dependency problem is if the growth rate itself continues to grow in a grand human Ponzi scheme.

To make this point clear I have poached the data from a great study way back in 1999 by Peter McDonald and Rebecca Kippen. They simulate a number of Australian population scenarios that represent some of the political views at the time ranging from Harry Recher’s view of a one-child policy coupled with zero immigration, to Tim Flannery’s view that a sustainable long term population target is 12million, to Jeff Kennett’s view that immigration can solve the coming uptick in the dependency ratio.

I show in the graph below the dependency ratio[1] based on the various population projections in their simulations (final populations in 2100 are in brackets).

A few things should be noted.

First, the lowest population projection, the Recher model, gets Australia’s population to 5million by the end of the century and reaches a peak dependency ratio (DR) of about 2.2. The highest population projection, the Kennett immigration solution, reaches a population of 929.5million (yes, a billion) in 2100, relying in a population growth rate of over 4% to keep the DR at it’s 1998 level of about 0.7.

In between these two extremes we have population paths that lead to populations between 12 and 50million by the end of the century, all of which result in a DR between 1 and 1.5 by this measure.

But here’s the thing. That 0.5 difference in the DR for what are radically high and low population projections can be totally offset by changing the retirement age by just one year - shifting the population at age 66 from dependent to working age.

At the moment the Australia pension age is shifting two years - from age 65 to 67. If social norms of employment change to accompany this, than any ageing problem is already solved.

In short, what seem like insurmountable demographic shifts are actually relatively slow and minor changes in economic terms. Not only does a declining youth dependency ratio offset much of the increase in the age-dependency ratio, but from the perspective of the economy as a whole the potential costs of ageing are minor compared the economic and environmental costs associated with rapid population growth necessary to suppress this ratio.

[1] In these scenarios the dependency ratio is weighted so that a child accounts for 3/4 of a person, and a retired older person (above age 60) accounts for 5/4 of a person.

Slight mistake above. I don't know what I was doing with my calculations, but around 1% of people are age 65 at the moment, meaning that shifting the retirement age up a year has a much smaller impact than I first estimated. I have an updated graph below that shows the effect of shifting the retirement age from 65-67, which is current policy in Australia. My point holds that if there is a dependency ratio problem from ageing, we have solved it better by this single policy than what we could under anything but the most extreme population Ponzi (green line in the simulations below, with a population of a billion in 2100).

Sunday, October 25, 2015

Queensland will ignore better planning

I wrote a submission during the consultation on reforms to the Queensland planning system. As you are probably aware, my research in this area focusses mostly on teasing out statistically the amount of favouritism happening in high value rezoning decisions.

My submission also focuses on the scope within the proposed Planning Bill for continued favouritism. But I also raise the point that we should enshrine at the highest level that governments of all levels, from Councils upwards, be able to capture the land value created by their planning decisions. As it stands the Bill suggests the opposite should be the case - that Councils are liable for compensation should they downgrade the value of land uses at a location (subject to the land being currently used for that purpose).

In no other area of government do we give away property rights for free, so often. It's a multi-billion dollar annual ritual of gift-giving from the public to politicly-connected landowners.

If you want to listen to a terrific podcast covering the types of mechanisms available to recover value gains from government policy I recommend last week’s Renegade Economist episode, which is the source of the below image as well.

You can download my submission here.

Saturday, October 24, 2015

Explaining everything explains nothing: Economics

The below comic probably hits a little close to home for most economists. Poking fun of economists’ naive attachment to their particular brand of rationality, and it’s immense body of hidden assumptions, usually gains responses that are merely signals of tribal loyalties and ignore the substance of the critique.

My colleague Vera te Velde made an improvement on the comic which is also included below, and my reading is that she wants to gain loyalty from the tribe more than defend rationality assumptions in economic models.

While it is certainly true that her improvement is consistent with core economic theories, it really highlights in my mind that these theories are essentially vacuous if they can explain anything and everything imaginable.

In response to the ability of other economists to fit everything within their model (meaning nothing can be excluded), simply because of creatively reversing out a set of preferences that explain the observed behaviour (or as Vera concludes De gustibus non eat disputandum), I wrote
There was simply no phenomena he could not explain with his beloved utility theory. But if the theory explains everything imaginable, then it predicts nothing. I resolved that the theory as it stands, in the revealed preference form, was not falsifiable, though he didn’t seem to understand why that would matter.  
Sure, humans often make calculated decisions, but the more I learn about the nexus between individual behaviour and how we behave in groups, the more I see very little value in rational-individualist views of economic systems that see all behaviour arising from God-given personal tastes. Without acknowledging the necessity of group-coordination mechanisms intrinsic in our behaviour, we are missing the main story.