Log into your member account to listen to this article. Not a member? Join the herd.

Two weeks ago, I waded into an emotive Twitter exchange on the subject of unpaid internships for university graduates. It quickly became evident that this was not in fact a debate but the outpouring of pent-up frustrations that is all too common among the youth generally – and more so among the better educated – that was the subject of two recent eReview articles (Hustler Nation: Jobless youth, millennial angst and the political economy of underachievement; Education, Social Mobility and the Enclave Economy: Revisiting the Kenya Scenarios Project).

Having jumped into the fray, seemingly on the side of unpaid internships, I found myself on the receiving end, perhaps deservedly so. Still, I felt that the opportunity to inject some sobriety into the debate was not to be missed.

I will dispense first with a comment on why compulsory remuneration of internships is bad economics.

Hiring workers is costly, and firing is costlier still, both financially and emotionally. The human resource practice has developed various methods to minimise the risk and cost of unsuitable hires, such as screening, references, rigorous interviews, psychometric evaluations, probation, (which minimises the cost of separation), outsourcing (which shifts the transaction costs to employment agencies) and the poaching of proven performers from rivals at a considerable premium.

Internships can serve the same purpose. Consider a firm that employs university graduates and is willing to offer a salary of Sh30,000. If it employs one who turns out to be unsuitable, it incurs a month’s salary and a recruitment cost of a similar amount, a loss of Sh60,000.

Alternatively, the firm can offer three two-month internships and pay an allowance of Sh5,000 a month, at the end of which it offers the most suitable intern a job. It is not hard to see why firms would resort to this strategy in these days when the quality, and even the authenticity, of academic qualifications has become very uncertain. It is also readily apparent from this example that firms using interns as a search strategy ought to pay them. But it would be a mistake, and probably counterproductive, to generalise.

Other firms may be willing to take on interns as part of their corporate social responsibility but may not have the budget for it. And there are also graduates out there who are willing to do unpaid internships. There are also entrepreneurs who could offer valuable internships but may not be able to afford the cost. Think of an entrepreneur who is keeping her struggling technology start-up afloat by earning extra income, say by taking up a part-time teaching position that pays her Sh30,000. If, however, she could get interns willing to pay the amount, she would happily give up the part-time work and concentrate on the start-up. Compulsion to pay interns, whether through policy or through social pressure (such as the trending #payinterns), has the effect of reducing the supply of internships, which is maximised by allowing all three options – paid, unpaid and paying internships – and leaving the market to do the rest.

As I argued in Social Mobility and the Enclave Economy, today’s university graduate is a victim of a legacy of privilege from the halcyon days of matunda ya uhuru (fruits of independence) when it was an automatic ticket to high-status public sector jobs. All Bachelor of Arts graduates were automatically absorbed into the civil service as administrators. Those appointed as “Bwana DO” (District Officer) moved into a large bungalow previously occupied by a white man and were provided with a Land Rover and a bevy of “APs” (Administration Police) to do their bidding. Few university graduates joined the private sector, but those who did went straight to the top and enjoyed lifestyles that their peers in developed countries could only dream of. What Kenyans seem not to appreciate is that this status was not obtained on merit, and had no relationship whatsoever with the economic value of university graduates. It was a case of replacing white privilege with black privilege.

This transition was more or less complete by the late 70s. Since then, university graduates have simply been trickling down the system and displacing the less educated. In the 60s, a graduate was assured of a leadership role, in the 80s a professional and middle management position, in the 90s entry-level work, and, of course, today they are not guaranteed anything all. Up until the 80s, it was inconceivable that a university graduate could be a police constable, but here we are.

In short, university graduates have continued to have expectations of entering employment at a fairly high level and climbing the socio-economic ladder as rapidly as previous cohorts did.

And there is another dynamic at play; intergenerational social mobility has been quite high in Kenya. For example, my grandparents were primary school-educated (quite a high level of education in the 1920s and 30s!). Their children, born in the 1940s and 50s, are mostly high school-educated with a few being university-educated, while all my siblings and most of my cousins born in the 60s and early 70s are university-educated. This has meant that we have, by and large, become accustomed to intergenerational social mobility. But now we are getting to the point where we have a critical mass of graduates whose parents are themselves university-educated and who are finding themselves a notch or two below their parents socio-economic status. This is bound to be stressful.

I came across a new policy document the other day titled the Draft National Automotive Policy. Its goal is to revamp the local motor vehicle assembly industry by way of import protection – what we call import substitution industrialisation. In plain English, this means the strangulation of the used motor vehicle import trade. Twenty years ago, I was involved in a long-running debate in the papers with one Gavin Bennet, then an industry spokesperson, on precisely this subject. At its peak in the late 80s, the industry produced 10,000 vehicles a year. Ownership was limited to institutions and the wealthy, while others had to wait for used cars to trickle down into the market. Choice was limited to less than ten models, and they were not particularly well-made. I argued that liberalisation would broaden vehicle ownership and create more jobs, that the economic benefits would more than offset the jobs that would be lost in the assembly industry. My prognosis carried the day; the cost of motor vehicles came down, variety increased with cars available for every pocket, with the attendant increase in employment in trade and services that we see today in the industry.

According to the policy document, the industry has ramped up its installed capacity from 28,000 to 34,000 vehicles a year. Production increased from a post-liberalisation average of 5,000-6,000 vehicles to 9,000 in 2015 and 2016 but it has since fallen back to the historical 5,000-6,000 range. It is unclear why an industry operating at 20 per cent capacity would increase its capacity. More importantly, it is readily apparent that the implementation of this policy would not stimulate new investment but rather, the utilisation of the existing plants and equipment.

The document goes on to claim that full capacity would create 150,000 jobs. This is balderdash. The same document acknowledges that at its peak, the industry employed 12,000 people, 3,000 directly and 9,000 downstream. If it were proportionate, a fourfold increase in production would translate to 36,000 jobs — but in reality, it would be less than proportionate. Unsurprisingly, the document does not factor in the jobs that would be eliminated in the second-hand imports and trade sector. All said, even 25,000 jobs would be a stretch.

In the intervening period, as globalisation was proceeding, China and India were ramping up motorcycle production, and driving down the cost. Before liberalisation, the cheapest motorcycle would have cost a well-paid university graduate at least a year’s salary. Today you can check one out from the supermarket at about three times the monthly minimum wage. And, of course, the industry has exploded, with an estimated 1 million to 1.5 million boda bodas on the road, and the jobs created also being in the same range. We are informed by the document that there are already eight motorcycle assembly plants operating at 50 per cent capacity (more than double that of the car assemblers), as well as an unknown number of “makeshift/informal” assemblers. This industry has developed in a liberalised environment without the protectionism that is now being proposed for car assemblers.

It is worth noting that the makeshift/informal motorcycle assemblers are mentioned only in passing; we have jua kali operations out there assembling motorcycles, and a proposed industry policy that merely acknowledges their existence. The existence of backyard assemblers tells us that motorcycle assembly requires little capital to set up. In economic lingo, assembling cars is capital intensive, while assembling motorcycles is labour intensive. The low capital requirements that make jua kali motorcycle assembly possible also mean that it is more easily scalable. The regional market for motorcycles is in the order of a million units a year. We already know that car assemblers cannot compete while the potential of an exporting motorcycle assembly industry is readily apparent. Also readily apparent is that this automotive policy is not about jobs but about returns on capital – profits – at the expense of jobs, competitiveness, equity and growth; that is, every important economic policy objective.

First, that it takes five times more capital to create a manufacturing job in Kenya than in India, and 50 percent more than in China, in other words Kenya’s industry is the most capital intensive of the three

The contribution of the owners of capital to our economic underachievement through policy capture – such as seen here – is under-appreciated. A World Bank study, Kenya Growth and Competitiveness Study, estimated our manufacturing productivity in the early 2000s at $3,500 per worker, vis-à-vis China’s $4,400 and India’s $3,400. Our capital (i.e. investment) per worker was estimated at $11,500, China’s at $7,800 and India’s at $2,400, translating to a capital output ratio of 3.3 to India’s 0.7 and China’s 1.74. What exactly are these parameters telling us?

First, that it takes five times more capital to create a manufacturing job in Kenya than in India, and 50 per cent more than in China. In other words, Kenya’s industry is the most capital intensive of the three. Second, that it takes $3.3 dollars of investment to produce a dollar of industrial output in Kenya, less than two dollars in China and less than a dollar in India. Moreover, the same study found that our labour was also the most costly, at $100 per month for unskilled factory workers, against China’s at $85 and India’s at $50. Let us put this into perspective: the amount of investment that creates one job in Kenya would create four jobs in India, and two in China. Moreover, Kenya is the least attractive investment destination as, of the three, it has the highest labour cost, a feature of the protectionist high-profit economy that the proposed motor vehicle policy is designed to restore.

More fundamentally, with close to a million young people joining the workforce a year, should we be protecting industries that require a million shillings to create a job, while the same investment can create five or more jobs elsewhere

We are churning out 150,000 university graduates a year. Let us say that this motor vehicle policy was implemented and indeed did create 25,000 jobs. Let us assume, generously, that a fifth – 5,000 jobs that is – were new university graduate jobs. How many sectors and industries would we need to protect to absorb, say, half the annual supply of university graduates? More fundamentally, with close to a million young people joining the workforce each year, should we be protecting industries that require a million shillings to create a job while the same investment could create five or more jobs elsewhere?

Last week the President took the occasion of the State of the Nation address to announce a financial scheme fronted by his family’s bank. Fish rots from the head.

Why is this policy so fixated on waking up a corpse? Because the policy has been written by the assemblers themselves. In fact, all the industry players are listed in the document by name. This is highly unusual. Analysis of specific enterprises does inform policy, but this is usually contained in studies, memoranda and background papers, not in the final policy documents. Final policy documents of this nature should be neutral. But what does it matter? Last week the president took the occasion of the State of the Nation address to announce a financial scheme fronted by his family’s bank. Fish rots from the head.