Nobody Can Produce the Study Behind “Half of Graduates Are Unemployable”

"Nearly half of Botswana graduates are unemployable." It appears in opinion columns, conference talks and policy framing. We went looking for the study. There isn’t one you can read.

The trail

The claim in its citable form — that nearly 50% of graduate respondents were still unemployed or economically inactive because of a mismatch between supply and demand — appears in Maunganidze, Faimau and Tapera, "Graduate Employability in Botswana: Challenges and Prospects", in the Mosenodi Journal, 2016.

That paper attributes it to two earlier studies: one by the Department of Vocational Education and Training in 2007, and one by the Botswana Training Authority in 2010. Crucially, it attributes both explicitly as "cited in Bolaane et al. 2010", which makes this a citation of a citation.

The first source has no record

The 2007 departmental study is named in the body of the Mosenodi paper and appears nowhere in its reference list. There is no author, no title, no publisher and no document. We could not establish that it exists in retrievable form.

The second source cannot be obtained

The 2010 study is cited properly: Bolaane, B. et al., Tracer Study on the Employment Outcomes of the Vocational Training Graduates, Gaborone, Botswana Training Authority. That is a real reference with real bibliographic detail.

Its only public copy that search engines still return, hosted by UNESCO-UNEVOC, no longer serves the document — the address resolves to a generic page carrying unrelated global statistics. The Botswana Training Authority itself no longer exists, having been succeeded by the Botswana Qualifications Authority. So the sample size, the fieldwork dates and the method behind the figure are all unknown.

And the paper carrying the claim is not nationally representative

Worth knowing on its own terms, because the paper is often cited as though it were a survey of Botswana graduates:

• 197 students surveyed at two institutions, the University of Botswana and Boitekanelo College, split 67.5% and 32.5%
• 167 graduates interviewed by telephone, recruited by snowball sampling because no graduate database existed
• One focus group of eight people
• Pilot conducted in November 2014

That is a legitimate qualitative study. It is not a measurement of a national proportion, and the paper does not claim to be one.

The word doing the most work

"Unemployable" is the term that travels, and it is not a finding. In the literature it functions as framing — "graduate unemployability" as a topic heading — never as a measured share. We found no study, Botswana or otherwise, that puts a percentage on graduates being unemployable.

What is actually on the record

Government’s own position is a mismatch claim without a number. The Permanent Secretary in the Ministry of Higher Education, Professor Richard Tabulawa, told the Parliamentary Standing Committee on Government Assurances in June 2026: "The jobs are there. In many cases, we do not have enough skilled people, or the skills available are not good enough to meet market demands."

That is a serious statement from a position that would know. It is also unquantified, and the absence is consistent: there is no national tracer study behind it.

Even the headline count is contested

The stock of unemployed graduates has two official values for effectively the same period. The Human Resource Development Council, working from the Statistics Botswana quarterly survey, reports 70,512 at the first quarter of 2024, against 48,076 at the end of 2019. The then Minister of Labour and Home Affairs, Anna Mokgethi, answering a parliamentary question in February 2024, gave 40,036.

Around thirty thousand people separate two government answers, and we found no published reconciliation. Two further precision points on the Council’s own series: its "46.7% increase over five years" compares the fourth quarter of 2019 with the first quarter of 2024, which is four years and one quarter, and the series does not rise smoothly — it runs 48,076, 57,632, 50,142, 62,515, 59,528, 70,512 across those observations.

Why we are publishing the absence

It would be easy to replace the discredited figure with a better-sourced one, and we cannot, because there isn’t one. The honest output of this check is a rule rather than a number: a claim about graduate employability in Botswana that does not name a dated, obtainable study is not evidence, however often it is repeated.

What can be sourced is set out in how many graduates Botswana produces and the skills Botswana says it needs. The other famous figure in this debate, the 4.6-month wait, is traceable but far narrower than its use implies: see where the 4.6 months comes from.