1. The Question
Quality of infrastructure is among the most quantifiable factors that connect macroeconomics with the microeconomics of how firms operate; and for Kenya’s electricity, it is especially so because of the survey of The World Bank Enterprise Survey in which they inquire from firms about the percent of time they experience outages and the percent of sales lost due to the outages. Kenya’s electricity supply has in fact become more reliable over the last ten years, there are fewer outages and lower sales losses, but firms in Kenya experience greater sales loss from outages than the average firm worldwide and in lower middle income countries.
This paper asks: what is the specific microeconomic mechanism through which electricity unreliability reduces Kenyan firm productivity, and does this mechanism fall equally on firms of different sizes, or does it help explain why Kenya’s firm-size distribution remains so skewed toward the very small and very large, with few firms in between?
2. The Microeconomic Mechanism
Power reliability impacts businesses in two separate ways, and each is related to firm size in some fashion that is relevant to the wider missing-middle challenge analyzed in other sections of this paper series.
Direct output loss during outages. As a result, there will be a stoppage in manufacturing, refrigeration, machinery, point-of-sale equipment, and lighting as long as the power failure is occurring. This is a simple case of loss of output costs, where the loss that is experienced by the company is measured in the form of the portion of sales that is lost due to power failures on an annual basis.
Self-insurance cost (the generator margin). Firms act rationally in dealing with risks of outages and invest in backup generation. However, it must be understood that a generator does not represent a free insurance against electricity outage, it entails costs in the form of initial investment, fuel (variable cost that is much higher than price of electricity per kilowatt-hour charged by the power utility, as it is very expensive to generate electricity from diesel fuel), and also maintenance expenses. Importantly, there are economies of scale involved in generator capacity, the more substantial a firm is, the smaller proportion of generator fixed costs it bears as a share of total production compared to a smaller firm, which is either uninsured or has a generator whose fixed cost constitutes a larger share of its revenue.
A compounding, size-regressive grid tariff structure. Separate from the outages, there is already a pricing structure to Kenya’s electricity rates whereby smaller commercial firms pay a higher per-kilowatt hour price than larger firms and firms in industry, the standard way for tiered pricing to work at utilities anywhere, although in the case of Kenya it means a second disadvantage in parallel with the first disadvantage noted above. It costs a smaller firm more to get the electricity to which it has access, and proportionally more to compensate for the electricity it does not have.
Overall, these are reasons to expect that unreliable and unfairly priced electricity will be a regressive cost relative to the size of the business, one that a bigger, more established firm can cope with and a smaller or micro firm cannot.
3. Evidence
Kenya’s outage-related sales losses remain above global and regional benchmarks despite improvement. In fact, as highlighted by the World Bank’s Kenya Enterprise Survey of 2018, the firms have experienced the loss of around 4% of their annual sales due to power outages, which is an improvement from the 6% recorded in the survey of 2013, but higher than the average of 3% losses recorded globally and specifically in lower-middle-income countries. The percentage of firms that have reported any outage in the period under review has also been reduced from 90% to 80%.
Firms are self-insuring at a rising rate, even as reliability improves. The percentage of Kenyan businesses having their own generator or being part of one increased from 57% in 2013 to 66% in 2018; this shows that there has been an increase in private capital investment in emergency power generation even though power outages at grid level have reduced. This conforms to the mechanism explained above where businesses will continue investing in insurance against their own risks of facing power outages.
The tariff structure itself is regressive with respect to firm size. In Kenya, the price paid by small firms for electricity is on average about 132% more than the price paid by large firms to access the same amount of electricity, a third premium included within the tariff rate itself and independent of any cost associated with electricity failure. This proves the second suggested channel above. Small firms have an already high cost of accessing electricity regardless of its generation.
Grid reliability and grid cost are linked through the same underlying vulnerability. The tariffs for electricity in Kenya have a variable component known as the Fuel Energy Cost Charge, which increases when more thermal electricity is generated using diesel generators - usually during drought conditions where there is less electricity being produced through hydropower plants, which used to be among the main sources of electricity production in Kenya along with geothermal energy sources. This implies that the same factors that increase the risk of outages also affect the cost of electricity available on the grid.
The scale of this problem globally underscores why it is a first-order productivity issue, not a minor operational nuisance. Based on microdata from the World Bank’s Enterprise Survey of more than 143,000 firms in 137 countries, the total annual cost of infrastructure disruptions (power, water, and transportation combined) for businesses has been estimated to be around USD 300 billion in the world, made up of utilization rate costs (USD 151 billion), sales losses due to electricity outages (USD 82 billion), and the extra cost of self-generation (USD 65 billion), thereby proving that the self-generation cost factor found in the Kenyan case is indeed a global factor.
4. Argument
This suggests the presence of unreliable and highly variable electricity pricing as a tool which serves to exacerbate the overall missing middle firm-size challenge faced by the country and discussed in other pieces of this series. The bigger and more established the Kenyan manufacturer is, the easier it is for it to allocate the fixed costs of its own backup generators to a substantial revenue base, and it pays relatively low tariffs for the commercial-industrial use of electricity. What the small or micro enterprise receives is the exact opposite in terms of both its baseline tariff rate and the fixed costs of a generator if available, and the outage cost as a percentage of revenues.
This is directly related to the capital-constraint process explored in this series’ youth entrepreneurship study where the growth-constrained small business would normally use their capital for expansion, employment generation, and investing in productivity-improving equipment, but instead divert it towards building generators to keep their existing production going amidst outages. Thus, electricity unreliability is not only a direct constraint on productivity, it is one particular, quantified avenue fighting for the exact same limited growth capital that this series’ other studies found to be the constraining factor for Kenyan SMEs beyond the micro stage.
5. Implications
- For infrastructure policy: Since self-generation costs constitute such a significant portion of the overall cost of unreliable power (about 40% of the world estimate noted above), investments in grid reliability have dual benefits since they not only help to save on the loss of output but also allow savings in privately owned capital spent on self-generating capacity, capital which otherwise could have been used to grow businesses.
- For SME-targeted policy: Since tariff design per se without any outages is a problem that is already putting small firms at a disadvantage by about a third unit, the matter of tariff design requires its own policy focus from that of reliability investment since both are compounding but different variables.
- For future primary research: A natural and appropriate next step for this research project is to extract the data on firm-size-disaggregated outage costs and generator ownership from the firm level microdata of the World Bank Enterprise Survey in order to examine empirically if there is any firm size regression in terms of the percentage of the sales that go towards outage costs in Kenya.
6. Limitations
This aspect of the size regressivity hypothesis in this paper comes from the combination of two different datasets, namely, the outage loss data from the country highlights in the Enterprise Survey, together with the comparison of tariff schedules in the utility pricing dataset, and not from the same analysis conducted at the firm level to test for size-specific differences in outage losses relative to sales in Kenya; the latter can easily be done using the microdata in the Enterprise Survey. The figure of USD 300 billion in infrastructure cost globally each year is a cross-country figure and is employed here merely to prove the point that the self-generation cost channel is an important driver of infrastructure cost and not a small part thereof, the Kenya-specific breakdown is unavailable. Lastly, the cause-and-effect argument that the higher cost of electricity causes the missing middle of firm size distribution (instead of being one of the factors that just coincidentally harms small firms) is an inference made from the above mechanism and evidence, but not a decompositional analysis of the effect of electricity costs relative to credit market constraints.
References (selected)
- World Bank Enterprise Surveys. Kenya 2018 Country Highlights and Kenya 2013 Country Profile.
- World Bank (2019). Underutilized Potential: The Business Costs of Unreliable Infrastructure in Developing Countries, global Enterprise Survey microdata analysis across 137 countries.
- World Bank Data. Value lost due to electrical outages (% of sales for affected firms) and Firms experiencing electrical outages (% of firms) indicators.
- Energy and Petroleum Regulatory Authority (EPRA), Kenya, commercial and domestic electricity tariff schedules.
- GlobalPetrolPrices.com (December 2025). Kenya electricity price comparison, residential vs. business, small vs. large business rates.
- Kenya Power and Lighting Company (KPLC), tariff category structure and Fuel Energy Cost Charge documentation.
Prepared as part of the East Africa Microeconomics Research Series. This is a student working paper that synthesises existing research and published data rather than presenting original fieldwork; feedback and data corrections are welcome via the contact page.