Last week one of my fellow columnists asked the question "What is Guyana's dependency ratio." This immediately sparked my curiosity as the dependency ratio is one of the basic demographic ratios. Being an actuary my day-to-day work involves projections using mortality, which is one of the key factors affecting the make-up of a population (the others being the birth rate and the net rate of migration). Thus the inputs and outputs of demographic studies can be of great use to actuaries. Unfortunately I was expecting the question to be rhetorical and I would find the answer within together with the source of the information, however, the article pretty much ended by posing the question asked in the title. I thus set out to answer the question myself and this week's column chronicles the journey in coming to it.
Gathering the data
Originally I planned to analyse the data from the 2002 census to determine the dependency ratio and then carry out a population projection to project its course into the future. It was with some trepidation that I approached the gathering of the data itself, given that data in Guyana is often only available in hard copy thus requiring extensive, time-consuming keying-in of data. I cannot emphasise how much productivity can be gained simply by using computerised systems which have the ability to efficiently transfer data between them.
Even hard copy can be difficult to come by as there is a substantial cost to its production. Given the old adage, knowledge is power, there are those who would prefer not to see information readily in the public domain. I firmly believe that a large driver of poverty is due to people making suboptimal economic decisions because they only have imperfect information, which, if they had the correct information to hand would not be made. Such is the importance of transparency of information (and the lack of) in Guyana that I plan to devote an entire column to it early into the New Year. Suffice to say I fully support the tabling of a Freedom of Information Bill and I was expecting a trying exercise in obtaining the census data.
Bureau of Statistics to the rescue
Imagine my pleasant surprise, when earlier this week while I was trying to find the most up-to-date inflation data in order to benchmark pay rises, after entering 'Guyana inflation' into Google, in the top hits was a link to a website for the Bureau of Statistics (http://www.statisticsguyana.gov.gy)! Even more pleasant was the wealth of information I found there - including a download containing a compressed folder with a wealth of demographic and economic data. Finally full census data was included plus a projection of Guyana's population for the period 2005 to 2025.
On downloading this document I must say that credit must be given to the Bureau of Statistics, not only for developing the website, but also packing it with relevant information. Provided the site receives updates regularly it will become a regularly visited site by any of those in the search of data on Guyana.
Population pyramid
The projection of population actually takes the form of a report 'Population Projections Guyana 2005-2025' by Sonkarley T. Beaie, MPhil; UNV-Demographer, Bureau of Statistics and is dated November 2006, so it is not surprising that Dr McDonald had not come across it when he prepared his piece last week.
The author should be commended for producing a paper which involves no small amount of work in its production. While I would like to have an opportunity to discuss some of the underlying assumptions made with the author, I do not think this is the forum to do so. In analysing the results below, it should be borne in mind that conclusions should only be drawn after verifying the underlying data and assumptions are appropriate.
The easiest way to see the characteristics of a population at a glance is the population pyramid, which graphs the number of people in by age-group and sex.
From this data the dependency ratio can be determined; in the case I have taken it to be the proportion of the population aged 15-64 to those aged outside that range. Thus at the last census the ratio was 1:1.5, that is, there were 1.5 people aged 15-64 for every person aged outside that range. While this is much better than Sub-Saharan Africa it is still much less than the 'sweet spot' described in Ian on Sunday last week.
An interesting feature of the population pyramid is the substantial group aged 0-14 which represents some 36% of the population. This may be due to under-reporting of ages or data collection problems, if not it appears Guyana has had a recent baby boom. If these children do not migrate as they come into adulthood this will lead to a substantial number of people moving into the working population in the next ten years or so.
Population projection
In order to quantify the impact of this effect on the dependency ratio it is necessary to project the population going forward. (To be consistent with other Caricom countries the position in 2000 has been estimated as the starting point.) The report includes projections under three sets of assumptions - high, medium and low birth rate. Under the medium and low birth-rate assumptions an improvement in mortality is also assumed. Under these assumptions the population moves as follows:
All three projections show an upward trend over the next 15 years or so. Under the low rate of birth assumption the net impact of migration is such that in later years the population will start to contract.
Dependency ratio
I calculated the dependency ratios for each of the three projections at each five-year period and the results are shown below:
Under all three scenarios the dependency ratio is expected to increase significantly over the next ten years or so - thus quantifying the effect I expected given the structure of the population pyramid. After that the ratio falls off with varying degrees of rapidity. Under the high rate of birth assumption the dependency ratio falls of quickest - this is to be expected given that more births equals more dependents in the early age groups, coupled with the impact of higher mortality, reducing the numbers in the working population.
All three scenarios show that if the data and assumptions underlying the report are valid, Guyana will face a massive increase in the working population in the next few years. The most immediate consequence is whether employment opportunities can be found for all of them.
