Skip to content

“Fiji’s elderly tsunami: the folly of ignoring ethnicity” (FT 8/2/2020)


Fiji’s elderly tsunami: The folly of ignoring ethnicity (FT 8/2/202)

 If a progressive Government plans to look after all of Fiji’s elderly equally on the grounds that “we are all Fijians”, does the “ethnicity” of the elderly matter?

Unfortunately, the answer is a very loud YES, especially when it comes to planning for the required homes for the elderly.

One of the facts of life in Fiji is that large segments of the population and the elderly do not eat beef or pork or even any meat at all, a preference deeply entwined with long held religious beliefs, culture and ethnicity.

Many of these Fiji citizens would even starve rather than eat their own food which has been cooked in pots used to cook the meats objected to.

Another fact of life, and not just in Fiji, is that the elderly very naturally prefer to live within their cultural groups.

This is no big deal when children go to multiracial schools, as they thankfully and increasingly do in modern Fiji.

But rarely do Indo-Fijian parents send their children to boarding schools where such “dietary or cultural mixing” may have to take place.

I suspect such dietary and cultural mixing will eventually come even for our elderly, but it is not going to be any time soon, and certainly not for large proportions of the massive numbers of aging Indo-Fijians (the tsunami) that is about to engulf Fiji.

Is the Fiji Government properly planning for caring after their culturally diverse elderly in the very near future, taking account of their special cultural needs?

I doubt it.

I doubt if even Indo-Fijian NGOs, communities or families are doing any such planning despite the trauma they will face in looking after their elderly as they age, develop dementia, and become bed-ridden.

It is an utter tragedy that the Fiji Bureau of Statistics, which should be in the forefront of warning the Fiji Government and Fiji public about the size of the impending crisis, is no longer releasing the ethnicity specific statistics which can show planners the serious nature of the coming elderly tsunami.

The end of FBS data on ethnicity

If you, the taxpayer, or even a parliamentarian were to ask the FBS: can you give us some statistics on the numbers of elderly by ethnicity going forwards ten years, the answer today would be: ‘We do not produce such data because “we are all Fijians” ’.

Yet they did produce such data once upon a time in the not so distant past.

From around 2012, no doubt as a political decision by the Bainimarama Government, the FBS stopped producing statistics on ethnicity.

None of their statistics released on the periodic Employment and Unemployment Surveys (EUS) every five years or the periodic Household Income and Expenditure Survey (HIES) (every five years) or even the most comprehensive data collection exercise in Fiji, the Census (every ten years, with the last being in 2017) has any table by ethnicity.

For any demographer attempting to project Fiji’s population into the future, not having data by ethnicity is totally irrational and crippling.

Fiji population projections need ethnicity

Here is a simple arithmetic question that illustrates the demographic problem.

If two cars are travelling away from home and their average speed is 75 km per hour, where will they be after one hour, if a helicopter is looking for them?

The simple “average” answer would be “75 km away”.

But one car could be travelling at 50 kph and one at 100 kph, and so a helicopter would have to look for one 50 km away, and one will be 100 km away.

BUT they might even be travelling in different directions away from home?

Similarly, Fiji happens to have two major ethnic groups whose proportions of the total population are totally different, and going in different directions.

Indo-Fijians, who were once more than 50% of the population, have had a much higher emigration rate since the 1987 coups; their women have had a lower and reducing fertility rate for 50 years (less than two children now, while indigenous Fijian women have a higher rate of three).   Indo-Fijian women have their first child at an earlier age (around 25) as opposed to 28 for indigenous Fijians who have their children later.

The net result is that Indo-Fijians are now around only 30% of the population and still falling while indigenous Fijians are more than 65% of the population and rising.

[Here is a hint of things to come:

My projections suggest that Indo-Fijians are today only 22% of school enrollments at Class 1, and that will decline further to 14% by 2027! Imagine the changes in enrollments at formerly “Indian” schools and colleges throughout Fiji. For example,  Indo-Fiians abroad may get  a shock to know what proportion of the school known previously as “Indian College” are still Indo-Fijians and what indigenous Fijians (and Chinese and Koreans for that matter).]

So you do not need to be a statistical demographer to understand that to project any group’s population into the future (a village, a town, a city or a country), you need to know

* the base population

* the births

* the deaths and

* the net emigration.

If the population is reasonably homogeneous – that is, made up of the same ethnic group – you can use national aggregates for these parameters, and your projections would be reasonably accurate.

But in Fiji you have two extremely different ethnic groups; their fertility and emigration rates have been totally different, and their base populations have been going in different directions (like the cars I referred to earlier).

So you cannot use national aggregates to project future populations.

Any decent demographer knows that you MUST dis-aggregate your projections separately by ethnicity.

That is exactly what was done by Fiji’s foremost demographer (Dr Martin Bakker) when he ran a workshop for the Fiji Bureau of Statistics demographers in June 2006 and produced a report with ethnicity parameters an essential part of the Report. See if  you can find it on the FBS website. Good luck.

He also produced the initial draft 2007 Census Analytical Reports which have also disappeared from public view.

However I did read a few early drafts, enough to do obtain more current demographic parameters for fertility and mortality by ethnicity, sufficient to do some reasonable projections.

Of course, more accurate projections can be done by the FBS – if they are allowed to do so by their political Lords and Masters, who have taken control of this once independent body.

My projections

Using the 2007 Census (which was the last one to give separate population numbers by ethnicity), I have done some population projections. I am using the publicly available PEOPLE software by inputting the following parameters by ethnicity:

* the base populations by five-year age groups (from the quite accurate 2007 census)

* assumptions on fertility going forward (indigenous Fijians around three children per woman of child-bearing age while Indo-Fijians with less than two, and both decreasing into the future)

* assumptions on mortality (life expectancy of Indo-Fijian women was above 70 while that for indigenous Fijian women was 68 or less)

 * assumptions on emigration based on the last figures supplied by the FBS to me (Indo-Fijians emigrating at around 5,000 per year since 1987 while indigenous Fijians are emigrating at around 1,000 per year).

Graph 1 gives you the population of Fiji based on census figures except for 2027 which is based on my projection using 2007 Census data.  You can see the huge change in direction for Indo-Fijians after 1987.

[This graph, and the brief period 1946 to 1986 when the dotted line (Indo-Fijians) was above the solid line (indigenous Fijians)  is also a “map” of the political turmoil in Fiji leading to the 1987 coup. The continued current steep rise for indigenous Fijians and the continued steep decline for Indo-Fijians, is also a sobering “map” of future politics in Fiji – for those that have eyes to see.]

Explosion of Indo-Fijian elderly

But what is not visible from the above aggregate projections, is that the proportions of those over 55 are going to change enormously, and differently for the different ethnic groups.

Graph 2 gives the Age Group 55 and over, as a proportion of the total population.

For indigenous Fijians it will rise from 10% in 2007 to about 13% now, and to 15% by 2027.

But for Indo-Fijians starting from just 12% in 2007, it is currently about 24% and will be about 31% in 2027.

This is a massive increase in the Indo-Fijian elderly as a proportion of their population.

But these proportions still do not give the true extent of the increase in burdens for the Indo-Fijian families – or for any family – since total populations also include children, who do not look after the elderly, and are themselves burdens on the family.

It is the people of working age in the household who take on the bulk of burdens of looking after the elderly.

Elderly as a percentage of working age people

Graph 3 gives the real extent of the elderly tsunami that will hit working people in all families. This measures the number of elderly as a percentage of working age people. In other words, how many working age people will be available to support each elderly person?

From about the same proportion in 2007 (23%), the proportion for indigenous Fijians is currently around 29% and will rise to 33% by 2027. This means that, for every elderly indigenous Fijian, there are three people of working age.

But the Indo-Fijian proportions will rise from the same 23% in 2007 to about 47% currently and will reach a massive 67% by 2027.

To put it simply, an Indo-Fijian couple of working age already have on average one elderly person they are looking after.

By 2027 the statistics show an average 1.4 persons per person of working age.

This means that many Indo-Fijian couples will have two elderly people to look after.

Some may even have four to look after, if parents from both the wife’s and husband’s side are living with them.

Remember also that in many families, the young couples have emigrated.

To give some perspective on the numbers, between 2007 and 2027 there will be an increase of some 81546 elderly people 55 years and over.

How many of these elderly people will families not be able to look after and will therefore need publicly provided “old people’s homes”?

If we assume a low 10% of families unable to look after their elderly, that implies places for 8155 elderly.

Assume we can squeeze 100 people into one old people’s home. That means 82 new old people’s homes of which 35 would need to be for Indo-Fijians.

Note that I am only looking at the changes from 2007 to 2027, not including those who are already at those ages. 

If you accept (as I did at the beginning of this article) that elderly Indo-Fijian people have particular cultural needs and preferences (diet, worship, etc.), how will the new old people’s homes manage this?

The numbers could be much higher as our family structures weaken further, as is already happening.

Are Fiji’s demographers warning the Fiji Government about this tragedy about to hit Fijian families?

Or does Fiji have no demographers left?

How exactly is Fiji currently dealing with the tsunami of the elderly, already crashing on Fiji’s shores, and about to become even more severe?

[In my next article, I give a rough overview of my extended family’s personal experience in looking after our elderly in Melbourne and Brisbane, and how there may be enormous employment opportunities developing in Fiji involving age care].



Comments are closed.

%d bloggers like this: