The curious case of Chinese visa grants in Australian higher education
Australia’s International student numbers are in the news again and recent data releases from the Department of Education (“DET”) and the Department of Home Affairs (“DHA”) have generated a flurry of media attention over the past fortnight.
Curiously, there are a number of recently published articles examining Chinese visa grant rates, declaring that demand has flat lined or that demand is under threat. Various commentators have then linked this trend to recent issues in Australian-Chinese relations that have been playing out in the media over the past year.
The claims around visa grant data did strike me as strange. When I examine enrolment and commencement data for Chinese students published by DET, all I see is continued strong growth from this cohort into the higher education sector.
Since I live in these data sets on an almost daily basis, I decided to have a closer look at recent Chinese visa grant rates to see what story this data was telling me.
Ultimately what I found is a number of concurrent stories may exist in the data depending on how you twist and turn the data sets. My lesson here is that the structure of the data is as important as the data itself.
Charting Chinese visa grant data.
For these dataset examination, I’m using the Department of Home Affairs Study in Australia Statistics, in particular the study visa applications granted pivot table published as at the end of January 2018.
I’ve set the parameters to Chinese nationals and primary applicants only. I’ve also set the sector to higher education* (aside: read the important point at the end regarding sector choice in this dataset).
First let’s examine the trend over full financial years.
The last completed financial year was 2016–2017. Between that and the previous period, growth in visa grants was 12% year on year for Chinese students wanting to enter the Australian higher education system.
However, many recent reports have been examining recent Chinese visa data on a half financial year basis. I can see the rationale, as we’re currently about half way through the current financial year and that permits looking at half year on half year growth.
This data projection does indicate that year on year growth in Chinese visa grants has indeed slowed, with the first sixth months of the current financial year showing only 4.2% year on year growth in Chinese visa grants. The previous period was 12.2%.
Let’s dig a tiny bit further, what is the trend on a month to month basis during this six-month window?
Notice anything curious?
Yup, there was a GIANT spike in visa grants in September 2016. Let’s expand the view of the data to a calendar year rather than a financial year and throw in a monthly average.
You can see that September 2016 visa grants for China was abnormally high. And the two months before for the calendar year 2016 were below average. Is that telling us a slightly different story?
You may recall that during the second half of 2016, there were wide widespread reports of DHA (or DIBP as it was known at the time) experiencing delays in visa processing due to the implementation of Simplified Student Visa Framework (“SSVF”) on July 1 2016.
The PIE reporting on visa delays.
Looking at the monthly trend above as well as knowing about delays in DHA, is it unreasonable to argue that there may have been a number of visas that would have been ordinarily granted in the previous financial year (2015–16) that happened to be delayed into the subsequent financial year (2016–17)?
If that were true, would that disrupt subsequent analysis of the visa grant data, making it difficult to compare year on year growth between 2016–17 and 2017–18 due to the way DHA publish their data? It would not be the first time that a change in visa processing rules has resulted statistical anomalies.
This may be a telling chart: when we look at visa grants over the calendar year cumulatively.
You’ll see that the 2016 grant trend starts to move in April and May, bottom out in July and August before catching up in September. This series crosses a financial year border in July of 2016.
Saying all that, it is difficult to tell from visa grant data alone whether what I speculate above is actually true or not, but I still think there is something important to keep in mind when looking at time series data — even in international education.
Ultimately, what I’m trying to say is that seasonality matters here.
If you look at the chart for visa grants in a calendar year above you will clearly see that the month of September 2016 is a clear outlier. If you know your time series analysis, you will know that outliers may impact the quality of analysis as well as forecasts.
Here’s some relevant stats-heavy links if you want to learn more.
Timeline bugbears!
As you may have guessed, I do get annoyed that DHA publish their student visa data sets is that they report in financial years.
This makes the data less useful to industry practitioners as many (if not most) education providers here in Australia would rarely think about overseas demand in terms of financial years. In general, we’re orientated to think in calendar years.
And when you map out the Chinese visa grant data in full calendar years, this story plays out with slightly less despair. Again, here’s a chart:
Looking at China visa grants in calendar years, we see year on year growth between 2016 and 2017 was just under 8%. The previous year growth was higher at around 12%, so growth rate has slowed down — but is that a cause for gloom and doom scenarios we’re reading about in the Australia media for the sector?
Your call.
(*) The sector may mean more than you think it does.
One final quick — and important point — about higher education visa grant statistics in the DHA datasets, originally raised by journalist Anton Crace in The PIE recently.
The sector a visa grant falls into depends on how the course is registered and not the provider. As a DHA spokesperson in Anton’s article confirmed: “Non-university education providers are able to deliver higher education courses.”
If you look at the TEQSA national register for higher education providers, you’ll note around 168 providers of higher education courses, of which 40 are universities.
If that is the case, I find it difficult to conclusively say that Chinese student demand for our universities is falling on the basis of visa grant data alone.
Feel free to comment below with any other perspectives, criticisms etc or just hit me up on twitter @dbfmurray.
