Economic Crisis, Income Gaps, Uncertainty, and Inter-regional Migration Responses: Kazakhstan 2000–2014

Abstract

There is ample empirical evidence that internal migration occurs in response to wage differentials; recently, evidence has emerged that international migration is deterred by rising destination uncertainty. However, to our knowledge, there has been no analysis of how internal migration responds to differing incentives during good times and bad. This paper provides insight into this issue using detailed regional economic and migration data for Kazakhstan during boom (2000–2007) and crisis (2008–2014) periods. While conventional forces are affirmed, we find that the crisis deters migration and weakens the effect of wage differentials – while also reducing the deterrent effect of relative uncertainty.

Acknowledgements

This paper has benefitted greatly from comments by Andrew Morrison, Juan Carlos Suarez Serrato, Paul Voss, and two anonymous referees. All remaining errors are our own. Data and code, additional tables, figures, and materials that accompany this paper are available in Online Appendix.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Useful antecedents include Hart (1975), Blomqvist (1978), Rogerson (1982), and Berninghaus and Seifort-Vogt (1991).

2. Concern has been expressed that economic data might be falsified or highly inaccurate. The former concern seems implausible: the pages of business journals and websites like www.forbes.kz or www.halykfin.kz abound with distressing news and critical stories about economic trends and policies. Given the strong formal sector and the incentives to register to receive social benefits, it is unlikely that permanent migration is greatly misstated. The situation regarding temporary and international migration is different. International Organization for Migration [IOM] (2015) estimates that Kazakhstan has roughly one million foreign residents, or about 6 per cent of the population. It cites a rough estimate that there are about 300,000 unregistered foreign workers and that the nation records about 100,000 ‘cases of breaches of the residence regime’ (p. 18); surely, many more go unreported. Thus, official data are unlikely to capture much and perhaps most short-term and international population movements.

3. Macro data are taken from the World Bank, http://data.worldbank.org/indicator/NY.GNP.PCAP.PP.CD/.

4. As elsewhere, global credit was squeezed: this mattered greatly for Kazakhstan, as its banks had engaged in a credit boom and had borrowed heavily from abroad. The IMF estimates that by 2008 external debt reached 44 per cent of GDP (see http://www.imf.org/external/pubs/ft/survey/so/2010/car081710a.htm), with these loans largely fuelling a private real estate boom and construction investment. The global recession also meant a decline in oil prices and, as oil accounted for about 60 per cent of exports, Kazakhstan’s exports fell from $76.4 billion in 2008 to $48.2 billion in 2009. However, Kazakhstan enjoyed tiny public debt and had significant reserves in its sovereign oil fund: it drew on this fund substantially to maintain aggregate demand and to soften the external shock. Details on the shock and its aftermath are provided by International Monetary Fund [IMF] (2014).

5. We also included ratios of real investment and real value of construction in our regressions. However, none of these variables was significant; they are available upon request.

6. The income gap ratio variable is a measure known as ‘severity of poverty’ and is defined as the average distance of those who are poor to a poverty floor measure.

7. www.damu.kz provides details on the national SME promotion program.

8. One should treat this last point lightly, since housing cost estimates are not based on conventional methods (repeat sales, as in Case-Schiller, or hedonic estimates) that control for compositional fluctuations and stock upgrading.