About Nick: i am an economist based in malaysia. I write about development economics, while sneaking in a pop culture reference or two.

Distributional Outcomes of a Falling Ringgit

Published in The Edge Malaysia, 7 - 13 December 2015.

In a previous article in this newspaper entitled, ‘Elasticity, Complexity and The Wire,’ I argued that whether or not our nation’s exports benefits from the weakening Ringgit depends heavily on their elasticity. That was, therefore, one caveat to the claim commonly propounded by some, including Minister of Tourism and Culture, Datuk Seri Nazri Aziz, that a falling Ringgit is good for exports and good for tourism. Certainly, ceteris peribus (holding all else constant), this would be true but we do live in a world where nothing is ever held constant except life and death.

Therefore, the answer to, “Is the falling Ringgit good for exports and tourism?” is, like all good answers, “It depends.” If this makes me a looney in the eyes of Datuk Seri Nazri Aziz, so be it though I think he should at least hear me out. Having argued that one of the factors that determine if a depreciating currency is good for exports and/or tourism is how inelastic our goods and services offered are, I would now like to argue for another potential caveat that should be added to the discussion. That caveat is the distributional outcomes that come from a depreciating currency.

To set the context, consider the following scenario. Suppose Malaysia’s GDP grew by a remarkable and totally unforeseen 10% in 2016 (unlikely to happen, but let’s take the supposition for the point of the argument). Is that good? Well, if you cared about the pure aggregate of things, yes, a 10% GDP growth rate is good. However, if you care about deeper things such as how that growth is distributed, would you say that the growth rate is good if only the top 1% of all Malaysians benefited from the increased economic pie? Perhaps, if you believe in (the now debunked) supply-side trickle-down economics. But I do think many of us would say that that situation is hardly ideal. Therefore, what is of paramount importance to understand is that what is good at the aggregate is not necessarily good when disaggregated.

Against that backdrop, we can then ask the following question – What is the distributional outcome of a falling Ringgit? Put another way, does the falling Ringgit impact the lower-income households differently vis-à-vis middle-income households or high-income households? These are salient questions to consider, as they would help to guide public policy response in the face of currency depreciation, the causes of which are largely out of Malaysia’s hands (i.e. global oil prices, a Federal Reserve rate hike and whatever it is that China wants to do).

To my knowledge, there has not been a study of this sort done in Malaysia and, indeed, in most places. Certainly, if the government has done this sort of study internally, I would be much obliged if they could release it for public consumption. However, until then, I will just assume that there is no such study done for Malaysia. Looking elsewhere for some guidance – if not on the results of the study but on the methodology to conduct the study (so we can then conduct our own study) – I turn to a recent (2015) study, entitled, “The Distributional Consequences of Large Devaluations” by Javier Cravino and Andrei Levchenko, economists at the University of Michigan.

In their study, they note that there is limited research on whether exchange rate changes impact different types of households differently and thus, they seek to fill that gap in the literature. To do this, they use micro level data (household level) – which, I must note as a ‘fiduciary’ responsibility, the Malaysian Department of Statistics is notoriously disinclined to share with the public (despite using public funds to conduct these surveys and/or censuses) – to observe household expenditures before and after the 1994 Mexican peso crisis, otherwise known as the Tequila Crisis.

To get a detailed understanding of the situation, the authors differentiate between two types of expenditures in household consumption baskets, which they label ‘across’ and ‘within.’ According to them, ‘across’ is defined as goods and services which are consumed by all types of households. These include food, haircuts, toiletries and so on. They found that, in Mexico, across product categories, low-income households spend relatively more on tradeables (such as food) while high-income households spend relatively more on non-tradeables (such as personal services). This may also be true in Malaysia, if we import a lot of our food, which we do.

‘Within’ is defined as differences in the consumption of goods and services that are linked to being part of specific household categories. One way to think of this is that low-income households spend relatively more on lower-end goods purchased from lower-end retail outlets while high-income households spend relatively more on higher-end goods purchased from higher-end retail outlets.

With these definitions in mind, the authors found that consumers in the bottom tenth of the Mexican income distribution experienced increases in costs of living that were 1.25 times larger than consumers in the top tenth. What compounds this finding is that these increases in costs of living were declining systematically moving up the income distribution. This means that the bottom tenth faced increased costs of living over all other deciles of the income distribution, while, for example, those at the 50th percentile faced higher increases in costs of living than those above them but not those below them.

Another important result that this paper highlights is that ‘within’ income distributions, inflation for less expensive product categories was higher than that for more expensive product categories. Indeed, there is empirical evidence that shows that exchange rate pass-through is lower for high-quality products. Thus, the exchange rate devaluation fell again more heavily on the lower-income households. All in all, the authors estimate that households in the bottom decile of the Mexican income distribution experienced increases in the cost of living between 1.46 and 1.6 times higher than the households in the top decile in the two years that follow the devaluation. This is, as they argue, largely due to the fact that “the poor spend a higher fraction of their income on tradeable product categories, and among tradeables, on categories with a systematically lower non-tradeable component.”

Turning back to Malaysia, this study on distributional outcomes on exchange rate changes on Mexican households is a timely reminder that what seems good at the aggregate may not necessarily be so when disaggregated. We need to understand the distributional outcomes of such major exchange rate movements on different categories of households. As the authors of the study above do, we can consider ‘within’ and ‘across’ goods and services consumption of households as a starting point. Therefore, it really is not that looney to say, “Well, maybe not…” to the question, “Is a falling Ringgit good for exports and tourism?” Indeed, it may be good, responsible public policy.

2016: Our Collective Obligation

The Myth of the ‘Sharing Economy’