ECONOMIC CRISIS AND INCOME DISTRIBUTION

It is often said that globalization brings risks as well as opportunities at the macroeco­nomic level. Greater integration with the global economy can lead to the economy being buffeted by global fluctuations in trade and capital flows.

...moving from the 25th to the 75th percentile in trade openness is associated with an increase in aggregate volatility of about 17.3% of the average aggregate variance observed in the data. The impact of openness on volatility varies a great deal depending on country characteristics, how­ever. For instance, we estimate that an identical change in trade openness is accompanied by an increase in aggregate volatility that is five times higher in the average developing country com­pared to the average developed country. Lastly, we estimate how the impact of trade changes across decades. It turns out that all three channels, as well as the overall effect, increase in impor­tance over time: the impact of the same trade opening on aggregate volatility in the 1990s is double what it was in the 1970s.

However, a major focus of the last two decades has been volatility and crises induced by financial flows. Financial crises appear to be the new normal in the global economy. Fully fledged global crises, such as the one that occurred in 2008-2009, or the East Asian finan­cial crisis of 1997, which also had global repercussions, are recognized as at least aided by the far greater ease of movement of portfolio capital around the world in the wake of capital account liberalizations from the 1990s onward. These global crises also have impli­cations for national level macroeconomic volatility, which has also been affected by trade openness. Indeed, Hnatkovska and Loayza (2013) argue that the increased volatility can be attributed more to crises (“large recessions”) than to the normal economic cycle.

There is now a consensus that volatility is associated with lower growth—Hnatkovska and Loayza (2013) present only the most recent assessment in this vein. However, this section will review the recent discourse on the consequences of economic crisis for the distribution of income—for poverty and for inequality.^[340] The literature has set out a range of channels through which a global collapse of the type seen in 2008-2009, or the more limited contagion effects of the crisis in 1997, feeds through into income distribution. Atkinson and Morelli (2011) and Baldacci et al. (2002) highlight the following channels:

1. Economic slowdown. As a “balance sheet adjustment” recession takes hold in orig­inating countries, it is transmitted through trade to other countries. Thus, each coun­try faces an economic slowdown. There is unemployment in the formal sector and consequent downward pressure on earnings in the informal sector. We would expect the impact of economic slowdown to be rising poverty and also rising inequality.

2. Relative prices and sectoral effects. For a particular country, the decline in interna­tional demand may be concentrated in specific sectors, with quantity and price effects.

14

3. Asset effects. Changes in interest rates and revaluation of assets can affect incomes and wealth at the top of the income distribution. Ifthere are major downward valuations and reductions in income from capital, then crises could reduce wealth and income inequality through this channel.

4. Policy responses. This includes the consequences of fiscal retrenchment, which will have impacts at the lower tail of the income distribution, or bank bailouts, which will affect the top end of the distribution. In general, fiscal retrenchment through reducing public employment, or support for public works schemes and other forms of unem­ployment support, would increase poverty and inequality in the social sectors. Bank bailouts would support asset values and incomes at the top end of the income distri­bution and increase inequality. Finally, an important channel linking crises and dis­tribution is the drastic devaluation most often undertaken as a response to a balance of payment crisis. This is equivalent to a drop in real wages and an increase in profits.

Each of these channels can have multiple impacts on poverty and inequality, so the overall effect is an empirical question. Ravallion and Chen (2009) focus on the 2008 global financial crisis and provide projections of the likely impact on poverty. They estimate that “the crisis will add 64 million people to the population living under a dollar a day.” The methodology for doing this, however, assumes no distributional change within a country, based on the observed regularity that “relative inequality falls about as often as it rises during aggregate economic contractions, with zero change on average.” Thus, Ravallion and Chen (2009) simply apply projected contraction in total consump­tion and assume this contraction to be distributionally neutral.

An attempt at identifying the impacts of poverty and inequality through cross-country regression techniques is presented by Baldacci et al. (2002). They define crisis episodes, identify appropriate controls of country-time spells, and estimate the impact of crises on different dimensions of income distribution. Not surprisingly, they find that crises are asso­ciated with rising poverty. However, in terms of income distribution, they find that “The main losers in terms of changes in income shares are not the poorest (lowest income quin­tile) but those in the second (lowest) income quintile. The income share of the highest quintile also falls in crisis years relative to pre crisis years.” Thus, treating this regression find­ing as a representation of the average outcome, the results are consistent with the assump­tions of Chen and Ravallion who state that crises are, on average, distribution neutral.

The post-1997 crisis experience highlights the country-specific differences that can arise. Hagen (2007) argues that income inequality rose significantly in Korea after the crisis. Similarly, inequality rose in Singapore and Malaysia, but it fell in Indonesia and in Mauritius (Atkinson and Morelli, 2011). Atkinson and Morelli (2011) assess the asso­ciation between crises and inequality for a large number of crises over a long period of time. They distinguish between banking crises and crises of collapse in consumption. They look at the time path of inequality on either side of the identified crisis. For the former, they conclude that “the empirical evidence suggests that cases in which inequality tend to increase following the crisis are in majority, although we should caution that the sample size is too limited to draw firm conclusions.” For the latter, “empirical evidence concerning ‘change in direction’ suggests that consumption crises are more associated with reduction in inequality.

It would seem, therefore, that no easy generalizations are available for the impact of crises on inequality, as might be expected from the multiple channels through which they can work and how initial conditions in a country can affect the impact. What this means is that we need country-specific modeling to analyze and to predict the impact of crisis on inequality. One such approach is that of a microsimulation model, as in the work of Habib et al. (2010). This approach combines macroeconomic projections with transmis­sion mechanisms to the income distribution:

The model focuses on labor markets and migration as transmission mechanisms and allows for two types of shocks: shocks to labor income, modeled as employment shocks, earnings shocks or a combination of both; and shocks to non-labor income, modeled as a shock to remittances. Shocks can be positive or negative depending on the trends outlined by the macroeconomic projections. In most cases labor income and remittances account for at least 75-80% of household income. (p-5)

Such country-specific analysis can then be used both to identify early warning indicators and to design possible policy responses. For example, the authors apply the model to Bangladesh and recommend monitoring of remittances and wages by sector as indica­tors of the need for action.^[341] A range of these models and methods is surveyed in BourguignonandBussolo (2012), and in Bourguignon etal. (2008). However, animpor- tant question arises as to whether we use anonymous distributions before and after crises or whether we use panel data, which follow individuals from before the crisis to after. Then, anonymous distributions can show no change even when there is considerable “churning” as a result of the crisis as pointed out by Robilliard et al. (2008).

15

20.5.