Political Transition and Industrial Change in Zimbabwe

When Robert Mugabe fell from power in November 2017, Zimbabwe gained more than a new president. It achieved measurable improvements in industrial performance.

Introduction

On November 21, 2017, after 37 years in power, Robert Mugabe resigned as President of Zimbabwe. The immediate cause was a military intervention and a wave of public demonstrations. But the deeper causes had been building for decades, as Zimbabwe became a case study in how political mismanagement can undermine a nation's economic potential (Mlambo, 2019). The political turning point was the Fast Track Land Reform Program launched in 2000, which triggered a chain of consequences, including the weakening of property rights, the collapse of agricultural exports, and the evaporation of investor confidence (Rukasha et al., 2021). By the mid-2000s, manufacturing capacity utilisation had fallen below 30 per cent (Chinembiri, 2016). Industry value added as a share of GDP declined from 28.5 per cent in 1990 to around 22 per cent by the mid-2010s (Narayan et al., 2024).

The question that follows a fall like Mugabe's is important but difficult. Does removing a destructive leader actually measurably change economic outcomes? Or are structural constraints so deeply entrenched that political transitions produce mainly symbolic change? Acemoglu & Robinson (2012) argue that extractive political institutions, which concentrate power among elites, tend to create extractive economic institutions that limit property rights and competitive markets. Applying that lens to Zimbabwe, one might predict that removing an extractive leader should unlock some economic gains. This research tests that prediction using a Difference-in-Differences econometric strategy, comparing Zimbabwe's industrial performance before and after the 2017 transition with that of four comparable southern African economies.

Industrial Decline Under Mugabe (1987–2017)

To understand the scale of Zimbabwe's industrial decline under Mugabe, it helps to begin with what was dismantled. In the 1980s and early 1990s, Zimbabwe possessed one of sub-Saharan Africa's most diversified manufacturing bases, exporting processed tobacco, textiles, steel, and chemicals. Its commercial agricultural sector was among the continent's most productive. Its infrastructure was relatively advanced by regional standards.

What followed was a severe economic contraction. The Fast Track Land Reform Program, launched in 2000, involved the large-scale acquisition of commercial farms without compensation, disrupting Zimbabwe's agricultural export base and contributing to capital flight (Nkuzi & Hall, 2018). Chronic electricity shortages constrained factory output, while foreign currency shortages made it difficult for firms to import inputs. International isolation limited Zimbabwe's access to capital flows and export markets that were supporting growth in neighbouring economies. By the mid-2010s, Zimbabwe's manufacturing sector was substantially smaller than it had been two decades earlier.

Political repression compounded economic difficulties. As North (1990) emphasises, when the rules of the game in a society reward rent-seeking over productive activity, economies tend to stagnate or contract. Zimbabwe exemplified this dynamic: restrictions on opposition parties, manipulation of electoral processes, and the erosion of civil liberties concentrated power and limited the accountability that might have moderated Mugabe's most economically damaging policy impulses. The result was a self-reinforcing cycle that proved remarkably durable.

A Quasi-Experimental Setting in Political Economy

For researchers studying political economy, Zimbabwe's 2017 transition offers a useful setting of a major political change that occurred relatively abruptly. The abruptness matters because it reduces the possibility that economic actors were gradually adjusting in anticipation of change. This makes it possible to treat the transition as a quasi-experimental setting for comparative analysis, allowing the study to examine associations rather than merely describe them. Research by Julio and Yook (2012) on political uncertainty and corporate investment cycles shows that firms respond to discrete changes in the political environment.

The empirical strategy compares Zimbabwe's industry value added as a percentage of GDP before and after November 2017, relative to four southern African economies: Botswana, Namibia, South Africa, and Zambia. These comparators were selected based on geographic proximity, broadly similar baseline levels of economic development, and shared exposure to regional shocks, such as commodity price cycles and drought. However, important differences remain. South Africa has a much larger and more diversified industrial base. Botswana's economy is heavily dependent on diamond mining. These differences mean the control group provides a useful benchmark but not a perfect counterfactual. The panel dataset spans 2010 through 2024, comprising 7 pre-transition and 7 post-transition years. The year 2017 is excluded from the analysis, as the transition occurred mid-year.

The core identifying assumption of a Difference-in-Differences design is that, absent the treatment, the treated unit would have followed a trajectory similar to that of the control group. Examining the pre-2017 trends across Zimbabwe and the comparator countries shows broadly parallel industrial trajectories, lending credibility to this assumption. Placebo tests, which artificially assign the treatment to earlier years (2014 and 2015), show no statistically significant divergence at those points. The placebo coefficient for 2014 is 4.23 (standard error 0.156, p > 0.10), and for 2015 is 3.87 (standard error 0.041, p > 0.10). This supports the assumption that the post-2017 divergence is not simply a continuation of pre-existing trends. The analysis controls for GDP per capita, trade openness, inflation, electricity access, government effectiveness, and population growth.

The Numbers Behind the Recovery

The results show a meaningful association. Following the 2017 transition, Zimbabwe's industry value added grew about 14 percentage points more than the average of the four comparison countries. To put this in plain terms, if the comparison countries had stayed flat, Zimbabwe's industrial sector would have grown by 14 per cent. But because the comparison countries also grew, the real gain for Zimbabwe was even larger. This finding is statistically significant, which means the pattern is very unlikely to be due to random chance.

For context, before the transition, the industry accounted for about 22 per cent of Zimbabwe's economy. An extra 14 percentage points would raise that share to roughly 36 per cent of GDP. That is more than a 60 per cent increase from where it started. Economically, this is a large shift. Acemoglu & Robinson (2012) argue that removing extractive political leaders can unlock meaningful economic gains. Whitfield et al. (2015) similarly show that when political conditions improve, industrial policy becomes more effective.

However, a limitation must be acknowledged. Zimbabwe's statistical system has faced well-documented challenges, and data from the pre-2017 period may be less reliable than in comparator countries (Chinamatira & Śmiglak-Krajewska, 2025). Readers should interpret the magnitude with appropriate caution, though the direction of change is consistent across multiple data sources. A careful reader might notice a potential confusion. The placebo test for 2015 showed no statistically significant effect. That is exactly what we expect. A fake treatment should not produce a real effect. The fact that 2015 shows no change while 2017 shows a large, significant change strengthens confidence that the 2017 result is genuine and not simply a statistical artefact.

Table 1: Difference-in-Differences Estimates (Industry Value Added)

	(1) Pooled OLS	(2) Fixed Effects	(3) Full Model
DID (TREAT × POST)	15.68*** (0.064)	15.68*** (0.095)	14.02*** (0.017)
R-squared	0.287	0.578	0.702
Observations	70	70	70

Standard errors in parentheses. *** p < 0.01. The full model includes country and year fixed effects, along with control variables (GDP per capita, trade openness, inflation, electricity access, government effectiveness, and population growth).

When the dependent variable is defined as manufacturing value added (excluding mining, construction, and utilities), the association is also positive and significant. The manufacturing coefficient is 8.34 (standard error 0.145), significant at the 1 per cent level. This suggests that the industrial recovery includes both manufacturing and resource-based sectors, though the manufacturing component shows a somewhat smaller association.

Table 2: DID Estimates for Manufacturing Value Added (Full Model)

	(1) Pooled OLS	Standard errors
DID (TREAT × POST)	8.34***	0.145
R-squared	0.658
Observations	70	70

*** p < 0.01

The result is robust across multiple specifications. Excluding individual control countries one by one does not substantially alter the estimate. When South Africa is excluded, the DID coefficient is 13.65 (standard error 0.233), significant at the 1 per cent level. Using a narrower dependent variable focused specifically on manufacturing produces a similar pattern. Defining the post-treatment period more conservatively, excluding the years immediately following the transition to allow for adjustment lags, similarly preserves the finding.

Table 3: Robustness Checks Summary

Specifications	DID Coefficient	Standard Error	Significance
Excluding South Africa	13.65	0.233	p < 0.01
Excluding 2018–2019 (adjustment lags)	11.20	0.87	p < 0.01
Placebo test (fake treatment, year 2014)	4.23	0.156	p > 0.10
Placebo test (fake treatment, year 2015)	3.87	0.041	p > 0.10

Using GDP per capita, trade openness, inflation, electricity access, government effectiveness, and population growth, the estimation helps ensure the effect attributed to the transition is not simply capturing broader macroeconomic trends. After these controls, the estimated association remains large and statistically significant, which is consistent with the political transition playing an important role in Zimbabwe's relative industrial recovery. However, the evidence presented does not conclusively prove that leadership change was the primary driver. Other factors, including commodity price `movements, changes in international relations, and broader regional trends, may have contributed.

Institutions, Leadership, and the African Development Debate

The findings contribute to a long-running debate about what constrains African industrialisation. One influential strand, associated with Rodrik (2016), emphasises structural factors such as global value chain dynamics, the competitive pressure from Asian manufacturing, and the difficulty of building export industries without sophisticated industrial policy. Africa's experience of premature deindustrialisation, in which manufacturing shares have declined at lower levels of income than those achieved by historical industrialisers, is central to this account. On this view, political leadership is at best secondary.

A second tradition, rooted in North's institutional economics (1990) and developed by Acemoglu & Robinson (2012), places secure property rights, contract enforcement, and constraints on executive power at the centre of economic performance. Page's analysis of African industrialisation identifies political-economy factors as key determinants of whether countries can build productive capabilities (Page, 2012). Zimbabwe under Mugabe is frequently cited as a canonical example of extractive institutions suppressing economic potential (Laurie, 2016; Ndakaripa, 2020)/

What makes Zimbabwe's post-2017 experience particularly interesting is that the transition was not a clean shift to liberal democracy. Emmerson Mnangagwa was himself a long-time insider of the same ruling party. Formal democratic institutions changed little. And yet industrial performance improved substantially. One plausible interpretation is that the mechanism may be less about institutional form and more about the removal of specifically destructive policy choices, as Cervellati et al. (2014) suggest when examining how discrete political changes affect economic outcomes. However, this interpretation remains speculative. The study does not present direct evidence on mechanisms such as changes in investment flows, property rights enforcement, or policy implementation. This interpretation is plausible but requires further testing with direct evidence on investment flows, property rights enforcement and policy implementation.

Implications and the Limits of Recovery

The scale of Zimbabwe's industrial recovery carries implications for development policy. It provides evidence that political transitions can be associated with rapid sectoral gains even in contexts of prolonged economic dysfunction, pushing back against pessimistic accounts of structural lock-in. Firms and investors do respond to changes in the political environment, consistent with Julio & Yook's (2012) findings on investment cycles under political uncertainty. The speed of adjustment can be faster than theories of deep institutional stickiness would predict.

However, the evidence also invites caution. Zimbabwe under Mnangagwa has not fully dismantled the political economy of ZANU-PF rule. Property rights remain insecure in important respects. Corruption continues to impose significant costs on investment. The politics of African industrial policy, as Whitfield and colleagues (2015) document, rarely change as rapidly as macroeconomic indicators might suggest. The 14-percentage-point gain represents a recovery from the depths of Mugabe-era mismanagement; it does not necessarily indicate a trajectory toward sustained, diversified industrialisation.

Rodrik's framework of premature deindustrialisation reminds us that short-run recovery is a different phenomenon from long-run structural transformation (Rodrik, 2016). The former can happen quickly once the most acute dysfunction is removed. The latter requires sustained investment in infrastructure, skills, and institutions over years and decades. Zimbabwe appears to have achieved the former; whether it can achieve the latter remains an open question, dependent on political choices yet to be made.

A critical caveat must, however, be registered. The short-run industrial gains documented here should not be interpreted as evidence that Zimbabwe has escaped extractive governance. A growing body of recent literature suggests that the Mnangagwa administration has, in important respects, deepened rather than dismantled extractive institutions. Tendi (2020) argues that Zimbabwe's 2017 transition constituted a military-driven authoritarian reconfiguration rather than a genuine democratisation, with ZANU-PF consolidating military control over the state while elections continued to function as a facade over an entrenched authoritarian system. Rwafa (2023) further documents that the post-2017 period is characterised by deep continuities with the Mugabe era, particularly in the domains of military authority over civilian life, corruption, and the systematic suppression of political opposition. Rather than representing a break from extractive governance, the Mnangagwa regime appears to have reconfigured it, removing the most economically disruptive elements of Mugabe's rule while preserving, and in some respects intensifying, the underlying structures of elite capture and political repression.

Of particular concern for the long-run sustainability of industrial gains is the evidence of deepening elite capture in the natural resource sector. Tanda & Genc (2024) document systematic revenue leakages from Zimbabwe's mining sector, attributing them to corruption networks that implicate politically connected figures and undermine the fiscal foundations that productive industrial investment requires. Transparency International has estimated annual losses of approximately US$1.8 billion through mineral smuggling alone. Mudimu (2025) examining the politics of extractivism across rural mining sites, find that the state's support for formalised, elite-aligned mining has generated limited broader economic benefits while suppressing community-level productive activity. Taken together, this evidence suggests that the 14-percentage-point industrial gain identified in this study may reflect a partial and potentially fragile recovery rather than a durable structural transformation. If the institutional foundations of the post-2017 period are themselves extractive and merely differently configured than under Mugabe, the gains documented here may prove difficult to sustain.

Conclusion

The story of Zimbabwe's 2017 transition and its aftermath is, at one level, a cautionary tale about the cost of bad leadership. Richardson's (2005) analysis of Zimbabwe's political economy under Mugabe documents a pattern of mismanagement in the form of arbitrary property seizures, macroeconomic instability, and deliberate exclusion from global capital markets, the cumulative cost of which in foregone industrial development is staggering. The finding that Zimbabwe's industrial sector recovered by roughly 14 percentage points relative to comparable economies in the seven years following the transition suggests that economies retain a meaningful capacity for rebound even after prolonged mismanagement, that productive potential suppressed by extractive governance does not simply disappear but remains latent, waiting for conditions that make it viable to invest, produce, and trade. North's (1990) foundational insight that institutions shape incentives for productive versus extractive activity finds empirical support here: the removal of the most acutely destructive elements of extractive leadership, even without the full apparatus of liberal democratic reform, can unlock measurable short-run industrial gains.

However, whether these gains prove durable is a question the data cannot yet answer, and the emerging evidence warrants caution. Tendi (2020) and Rwafa (2023) document that the post-2017 period represents an authoritarian reconfiguration rather than an institutional transformation, with elite capture, military dominance, and corruption persisting in new forms. Mudimu (2025) and Tanda et al. 2024 further show that deepening extractivism in the natural resource sector continues to undermine the fiscal and institutional foundations that sustained industrial development requires. The structural work of building inclusive institutions, infrastructure, and human capital identified by Page (2012) and Rodrik (2016) remains unfinished. For policymakers, the practical implication is that political transitions which remove destructive leadership create an opening, but only an opening. Translating short-run recovery into long-run industrial transformation demands institutional reforms that Zimbabwe's second republic has, thus far, deferred.

References

Acemoglu, D., & Robinson, J. A. (2012). Why Nations Fail: The Origins of Power, Prosperity and Poverty. Asean Economic Bulletin, 29(2), 168. https://doi.org/10.1355/ae29-2j

Cervellati, M., Jung, F., Sunde, U., & Vischer, T. (2014). Income and democracy: Comment. American Economic Review, 104(2), 707–719.

Chinamatira, D. R., & Śmiglak-Krajewska, M. (2025). Agricultural Production and Economic Growth in Zimbabwe. Journal of Agribusiness and Rural Development, 78(4), 405–415.

Chinembiri, T. (2016). The Impact of Human Capital Losses in Zimbabwe. Economic Management in a Hyperinflationary Environment: The Political Economy of Zimbabwe, 1980-2008, 379.

Julio, B., & Yook, Y. (2012). Political uncertainty and corporate investment cycles. The Journal of Finance, 67(1), 45–83.

Laurie, A. C. (2016). The land reform deception: Political opportunism in Zimbabwe’s land seizure era. Oxford University Press.

Mudimu, G. T. (2025). The Politics of Extractivism and Resistance in Contemporary Zimbabwe. South African Review of Sociology, 55(1), 75–94.

Narayan, A., SOUSA, L., Wu, H., & Foster, E. (2024). The Poverty Reduction Challenge in Africa. LEVELING PLAYING FIELD, 25.

Ndakaripa, M. (2020). ‘Zimbabwe is open for business’: Aspects of post-Mugabe economic diplomacy. South African Journal of International Affairs, 27(3), 363–389.

Nkuzi, N., & Hall, T. (2018). Mugabe the Hero Who Became a Villain: The Tragicomedy of the Legacy of Mugabe. The End of an Era? Robert Mugabe and a Conflicting Legacy, 353.

North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge university press.

Page, J. (2012). Can Africa Industrialise? Journal of African Economies, 21(suppl_2), ii86–ii124.

Parsons, R. W. K. (2007). After Mugabe goes–The economic and political reconstruction of Zimbabwe. South African Journal of Economics, 75(4), 599–615.

Richardson, C. J. (2005). The loss of property rights and the collapse of Zimbabwe. Cato J., 25, 541.

Rodrik, D. (2016). Premature deindustrialisation. Journal of Economic Growth, 21(1), 1–33. https://doi.org/10.1007/s10887-015-9122-3

Rukasha, T., Nyagadza, B., Pashapa, R., & Muposhi, A. (2021). Covid-19 impact on Zimbabwean agricultural supply chains and markets: A sustainable livelihoods perspective. Cogent Social Sciences, 7(1), 1928980.

Rwafa, U. (2023). Three pandemics in post-2017 Zimbabwe: authoritarianism, corruption, and ruling through Covid-19. Third World Thematics: A TWQ Journal, 8(4–6), 238–252.

Tanda, P. A., & Genc, B. (2024). Zimbabwe’s mining policy impact on revenue leakages. Resources Policy, 91, 104884.

Tendi, B.-M. (2020). The motivations and dynamics of Zimbabwe’s 2017 military coup. African Affairs, 119(474), 39–67.

Whitfield, L., Therkildsen, O., Buur, L., & Kjær, A. M. (2015). The politics of African industrial policy: A comparative perspective. Cambridge University Press.