Sen on famines and democracy

by Daniel on January 6, 2005

This is by way of a followup to Chris’s comment on Nick Cohen’s article on the pointlessness of providing disaster relief to governments who don’t care about their citizens. I’ve never been a big fan of Sen’s dictum that “democracies don’t have famines” – I’ve always regarded it as being a slogan on a par with “no two countries which have a McDonalds have ever gone to war with each other”. I was originally just going to point out that the only African country which has managed to stay clear of famines entirely since independence is Kenya, which has not been a democracy and leave it at that, but I ended up looking up the original quote from Sen’s “Development as Freedom” and this ended up expanding somewhat into a more general piece on the subject of democracies and developmental states. I’m actually pretty sympathetic to Sen on most of the issues he writes about, and I hope readers will bear that in mind, because it is more or less impossible to resist making a few harsh remarks when you find out that the original quote, from page 16 of the paperback edition of Development as Freedom is, verbatim:

“It is not surprising that no famine has taken place in the history of the world in a functioning democracy – be it economically rich (as in Western Europe or North America) or relatively poor (as in postindependence India, or Botswana or Zimbabwe)”

Emphasis added, of course. I’d make it clear from the outset that I am not playing “pin the anachronism on the donkey” here; Development as Freedom was originally published in 1999, and the paperback edition I have is from 2001; there don’t appear to have been any changes made for the paperback but there could have been. Mugabe joined in the civil war in Congo in 1997, faced riots in 1998 and began stealing farms in 1999, the year in which the Movement for Democratic change was formed (for what it’s worth, the MDC was formed in September 1999 and the latest citation I can find in the book is to a newspaper article from June 1999). So it is clear that when Sen refers to Zimbabwe here, he is referring to Mugabe’s one-party state, and not to some immediate postcolonial Nirvana of the development studies literature. I’d emphasise, however, that this slogan only appears twice in the entire book and not at all in the two chapters which actually deal with democracy and famines. Almost all of the analysis below is actually cribbed from Sen’s book; I’m arguing against his slogan, not against him.

This ought to raise suspicions right from the get-go, of course. If the “democracies don’t have famines” factoid is a proposition that can have Zimbabwe advanced in its favour, then it is clear that it is a bit of a David Lewis gambit – a proposition which has been cunningly constructed so as to have no counterexamples. And a proposition like that is unlikely to bear as much empirical weight as one might want to place on a useful-sounding slogan like “democracies don’t have famines”. These suspicions are entirely justified, because it appears to me that Sen was entirely correct to include Zimbabwe on that list in 1999 and 2001 and (even more strongly) I suspect that if he was being consistent, he’d have to leave them on the list if he brought out a revised edition of Development as Freedom in 2005. Two simple reasons for this:

1. On the criterion Sen appears to be using, Zimbabwe is a democracy. Since the word “democracy” has been abused by the “democratic people’s republics” of the world (and indeed by the “liberal democracies”) so much as to have lost all meaning, it’s easy to forget that it can be used in a descriptive as well as a normative sense. Nobody but a fool or someone being contrary for the sake of it would argue that Mugabe’s Zimbabwe is anything other than a horrible despotism, or that what we normally regard as “democratic freedoms” are respected there. But … they do have elections. Horribly corrupt elections, but elections in which an opposition party stands (note that this isn’t a necessary condition for being a democracy for Sen; Zimbabwe was a one-party state between 1988 and 1999). An opposition party which has to be incredibly brave to stand up to the despicable treatment it receives from the government, but one which nevertheless wins seats, took Mugabe within an ace of losing the last Presidential election and actually beat him in local elections in 2003.

Zimbabwe even has a free press. Not a free press in the sense in which you or I would recognise it; journalists are regularly harassed both through the courts and through physical intimidation, and censorship is rife. But nonetheless, the censorship is not total, and the newspapers are not state-controlled in one important sense. That sense is the one alluded to in Nick Cohen’s article linked to by Chris; if there was a famine going on in Zimbabwe, the Zimbabwean press would print that there was one, and this is an important qualitative difference between them and, say, the Burmese press.

So, Zimbabwe is a democracy in the only objective sense which matters for Sen’s theory, which is that Mugabe can’t keep on ruling it if the people don’t want him to (and the same applies to his designated Zanu-PF successor). This was actually how Zimbabwe got into the current mess; Mugabe started grabbing farms not because he wanted to, but because he needed to do something for his political base after the Congo disaster. If there was a serious famine in Zimbabwe, Mugabe would not be able to hold on to power and he knows it.

[at this point, it is understandable to raise the issue of the disgusting practice of withholding food aid to political opponents, which Zanu-PF apparatchiks definitely did, and which there is decent evidence that government bodies directly controlled by Mugabe did in the 2002-03 food crisis. I would make two points here. First, there wasn’t a famine in Zimbabwe in 2002-03; it was averted (as it was in borderline democratic Malawi and Zambia, democratic Senegal, democratic but dysfunctional Kenya and entirely undemocratic Ethiopia and Eritrea). Second, the information that if a democratic government can identify a subset of its citizens that it doesn’t care about, it can starve them to death was already available to Sen; presumably he is aware of the Irish famine and does not regard it as a counterexample to his theory.]

and the second reason …

2. On the criterion Sen is using, Zimbabwe has never had a famine. No really. Zimbabwe has not had a famine since independence and has not had one in the last ten years; I will bet large sums of my own money on this proposition if the term “famine” is defined as it is in Development as Freedom. Zimbabwe has, since 1999, seen an economic collapse, a collapse in food production, an epidemic of malnutrition and a shocking increase in mortality rates. But it has never had a famine; an episode of mass deaths due to starvation rather than malnutrition. It looked like they were going to have one in 2002, but the UN’s World Food Programme and the Consortium for Southern Africa Food Emergency (an umbrella group of charities, since renamed the Consortium for Southern Africa Food Security) managed to avert it.

This second point is a rather important one, which goes to the heart of why I think “democracies don’t have famines” is a bad slogan. It sounds awfully like “the poor are better off under democracies”, but it doesn’t mean that at all (and indeed Sen spends two chapters of his book explaining that it doesn’t. It’s a statement about one particular kind of misfortune that can befall the poor, famines, and doesn’t directly say anything about malnutrition, AIDS, malaria, gender-selective infanticide, cholera or any other of the silent killers that account for the majority of excess deaths in poor countries. Peter Griffiths in his book “An Economist’s Tale“, which so help me I will review soon, suggests that “famine” is more of a television producer’s concept than a development economist’s one and there is a great deal of truth to this. Like tsunamis, famines are big dramatic events, and like tsunamis they are situations where a clear plan of action suggests itself, usually involving the organisation of the movement of lots of objects relative to the surface of the earth. For this reason, aid agencies and the man in the street is correct to focus on them for good Hayekian reasons; because these are situations where the diversion of resources from the market into a planned operation can clearly do identifiable good. Solving the problems of malnutrition and malaria would certainly save more lives than anything we can do for the tsunami victims, but a population’s state of being vulnerable to malnutrition and malaria is in general the result of complicated social arrangements which are extremely difficult to make plans about[1].

It is this “plannability” of famine mitigation which is at the heart of Sen’s argument about famines and democracy. The actual reason why “democracies don’t have famines”, according to Sen, is as simple as pie. Think about it this way; it would be a very bad famine indeed that killed even as much as 10% of the population. The 10% of your population who are vulnerable to a famine are usually the poorest 10%, so their consumption is often as little as 3% of your GDP. And the famine comes about when their consumption drops below subsistence level, not when it disappears entirely. So basically, it is reasonable to suppose that the famine-vulnerable population could be brought back to their original level of consumption by diverting as little as 1.5% of GDP, and a government which can’t divert one and a half per cent of a country’s output is not a government worthy of the name. So, Sen’s thesis about democracies and famines is based on the fact that it is actually very easy to prevent or mitigate a famine if you really want to. (It is an important irony that the way in which a democracy goes about preventing a famine is to start acting like a planned economy; interfering with the price mechanism, diverting resources and even commandeering private property. But this is only a contradiction for people who have made the mistake of wholly identifying “democracy” with “liberal democracy” and “liberal democracy” with “laissez-faire economy”. And I suspect that it would take only a couple of missed suppers to cure these people of their error).

The really interesting point is that it is only democratic governments which really have any interest in preventing famines. If you aren’t accountable to the population and only care about your own wealth, then you will not typically prevent famines. Why not? (I cribbed these from Peter Griffiths’ book, btw, which is sort of a novelisation of Development as Freedom and a much more exciting read)
1.A famine is usually the result of significant movements in the prices of important commodities. It is really quite easy to profit from these price movements if you are in control of the supply of the commodities and know that the famine is coming.
2.If you get early warning of a famine and start taking steps to prevent it, then the outside world will usually give you food aid in kind, to move through your distribution network until it reaches the needy. If, however, you wait until people are dying in front of the television cameras, then you will get aid in cash, in order to buy food locally (which is quicker than shipping it in from donor countries) and to set up the distribution network. Cash is easier to steal than rice.
3.If you are a genuinely undemocratic regime, then it is quite possible that even if you fundamentally like the people and don’t want them to starve, then nobody will tell you that there is a famine going on, because they are yes-men and scared of telling you the truth[2]. This was apparently always happening in Malawi; the leadership of the Women’s Movement of Malawi prevented at least one famine because they were the only group who felt able to tell the truth.

So, “democracies don’t have famines” should really read “there is no excuse for having a famine if you are a government which cares about its population and is in control of its territory”. This is less catchy, but it has two important advantages.

First, it doesn’t make any presumptions about democracies. Sen is certainly right to emphasise that, empirically, democracies are much more likely to care about the deaths of their citizens than any other kind of regimes (and are good for many, many other reasons; for people who want to peek at the end of “Development as Freedom”, he’s in favour of them both). But there are countries like Kenya which have a very good record in famine prevention and which haven’t been democracies for much of their history. If you start judging countries on the basis of a sweeping rule about “democracies” rather than on their actual record, then you’re doing the development equivalent of Coase’s “blackboard economics”, and you end up with nightmare ideas like “rights-based lending[3].

And second, by pointing out that there are two real issues here – the ability of the government to act as well as than its willingness – we draw attention back to an important caveat in Sen’s original formulation of the slogan which is not always preserved in quotation; that history does not provide examples of famines in functioning democracies. There are lots and lots of non-functioning or barely-functioning, kinda-sorta democracies in the world (Africa is full of them), and they have famines just like their neighbouring states. Since my cynical assessment is that the mission to the world which the neoconservative movement and its allies on the “decent left” have embarked upon is likely to leave us with a lot of non-functioning democracies which are not in physical control of the territory assigned to them on the map, we probably shouldn’t be surprised when they continue to have famines and should make plans to deal with them when they do.

Footnotes:
[1]Plain speaking bores who were just winding up for a long post about “Hah! The postmodernist fool Davies thinks that malaria is a social construction!” are cordially invited to go and have a wank instead. Malaria is a disease caused by an organism spread by mosquitoes. The condition of being vulnerable to malaria is a property possessed by some people and not others. The chief influences on whether someone has the property of being vulnerable to malaria are what kind of house they live in, where it is located and whether they have access to antimalarial measures like treated mosquito nets, DDT sprays and chloroquine. The factors which determine whether you live in a mud hut or a house with windows, and whether your pharmacist will give you chloroquine are in general related to your wealth and social status and these in turn are determined by social facts, not physical ones.
[2]Democracies can sometimes get the opposite problem; not so long ago, Senegal had to hand back a load of aid and apologise to the donor community after local media reports had got out of hand and led the government to believe that the country was facing famine when it wasn’t.
[3]This link, and the one in the introduction linked to the word “Zimbabwe”, are to previous CT posts by me on various subjects. I seem to remember that Timothy Burke added very useful comments on the opposing side to both of them and hope he will again.

{ 37 comments }

1

Matt 01.06.05 at 12:36 am

Dsquared,
I don’t have much of a substantial argument against your post, but am curious about one point- why focus on “Development as Freedom”? Because Cohen did? (I didn’t read his article.) I ask becuase of course Sen has been talking about this at least since the 70′s, and so it looked funny to me when you call the line from DasF the “original quote”. Do you mean the one Cohen quoted? And do you think that the earlier (and more technical, I believe) work Sen did on this has the same problems?

2

dsquared 01.06.05 at 12:46 am

Basically because DaF is the one that came out in a handy paperback edition so it’s the one I’ve got. I don’t think any of Sen’s work has these problems – all of the issues I’ve identified are right there in Sen and they are his ideas rather than mine. It’s just the slogan and its abuse by people with a particular slant on pushing “Democracies” that I’m wary of.

3

P O'Neill 01.06.05 at 1:27 am

When Sen does pop versions of his analysis (such as in the factoid), it can seem that there are two case studies driving the slogan: post-independence India compared with pre, and Ireland in 1847. But as DD says, the underlying proposition is surely that if you really want to prevent famines, you can, whereas in the Irish case, some combination of ignorance and ideology drove the lack of famine relief. One can imagine an 1847 Ireland with the same political structure within the UK (and thus “undemocratic” by Sen’s definition) but with a different set of policymakers in London, where the famine might have been better mitigated.

4

Sam 01.06.05 at 2:19 am

While Daniel’s qualifications seem reasonable enough, let’s not loose sight of the bigger picture here. It is the India-China comparison, post-1947/1949, that is perhaps most telling for Sen’s analysis, which provides a good first explanation for the greatest famine of the twentieth century, the Great Leap Forward. Millions and millions of Chinese farmers died because the undemocratic state refused to listen to them, or even to the Minister of Defense who spoke the truth to Mao, when they said they were starving, and the political system – even though it was premised on a peasant revolution come to power – did not respond. The failures in North Korea in recent years, where people have been dying in large numbers from starvation, would also be fairly well explained by Sen’s analysis. These are famines. And Sen does a good job in helping us understand them.

5

Sam 01.06.05 at 2:20 am

While Daniel’s qualifications seem reasonable enough, let’s not lose sight of the bigger picture here. It is the India-China comparison, post-1947/1949, that is perhaps most telling for Sen’s analysis, which provides a good first explanation for the greatest famine of the twentieth century, the Great Leap Forward. Millions and millions of Chinese farmers died because the undemocratic state refused to listen to them, or even to the Minister of Defense who spoke the truth to Mao, when they said they were starving, and the political system – even though it was premised on a peasant revolution come to power – did not respond. The failures in North Korea in recent years, where people have been dying in large numbers from starvation, would also be fairly well explained by Sen’s analysis. These are famines. And Sen does a good job in helping us understand them.

6

derek 01.06.05 at 2:55 am

As a liberal, I’m happy to turn Sen’s dictum around and look at it from the other end: no country in which there is a famine can be said to be a functioning democracy.

(the famine is either localised, in which case the country is functioning, but failing to function as a democracy, or truly universal, in which case the country is not functioning at all)

Expand “famine” into “poverty” and the USA looks less and less like a functioning democracy.

7

Lee Scoresby 01.06.05 at 2:56 am

Sam is dead on here. If I remember back to the days when I took a class with Sen, the main point is that democratic governments have institutional mechanisms that tend to prevent famines. That does not mean that authoritarian states *will* have famines, only that they lack an important structural check on the development of famines.

8

Lee Scoresby 01.06.05 at 3:04 am

Derek’s observation is interesting in another context. Another frequent argument about the virtues of democracy, is that they don’t tend to kill their own people in large numbers (RJ Rummel has an extensive website promoting his variant of the argument). Michael Mann argues, however, that this is a coding artifact – by the time states do kill large numbers of their people, they don’t look like liberal democracies. However, it may be the case that *democratization* creates conditions for mass violence, hence there may be a link between democracy and “democide.”

9

Dave F 01.06.05 at 8:21 am

Zimbabwe is a sham democracy and people are starving to prop up the dictator. The point is that dictators are not answerable to the people on what they do with their money or how they run the agriculture sector. However, Zambia is a democracy and does have a food problem (drought is no respecter of political piety).

10

Sebastian Holsclaw 01.06.05 at 8:51 am

“Expand “famine” into “poverty” and the USA looks less and less like a functioning democracy.”

That is quite an expansion. If you let me expand “gravity” into “electro-magnetic force” I have a fully functioning unified field theory. Furthermore you could quite easily argue that by world standards almost no one in the USA is in poverty. ;)

11

abb1 01.06.05 at 9:16 am

There are no democracies, they don’t exist. There are different mechanisms allowing the governed to express their dissatisfaction, mechanisms ranging from periodic opinion surveys (‘elections’) to popular uprisings. What does this have to do with famines? Not much, IMO.

12

Scott Martens 01.06.05 at 9:32 am

Sebastian has a point – although not necessarily the one he thinks he’s making. :^)

There is a certain circularity to the idea that functioning democracies don’t have famines. If it has a famine and it’s democratic, it is by definition not functioning. But the point is that a responsible government is one that is threatened by the existence of famine on its territory. In a functioning democracy, this threat comes through elections. In another state, it may be some other form of embarasment. However, the same state may not feel threatened by other sorts of outcomes that could be identified as failures. Clearly, it would be difficult for an American government to win an election following a famine; but equally clearly American governments are not threatened by geting involved in military quagmires or by promoting human rights abuses.

Some governments that routinely hold elections are threatened by widespread poverty and others aren’t. The existence of poverty does not appear to create the kind of shame and retribution that a famine does. This suggests that Sen’s conclusions are far less a universal rule than a question of values. If 90% of the population doesn’t care that 10% is dying of famine, democracy will not create the kind of threat necessary to produce action. Most people consider starvation of any segment of the population a transparent sign of failure. If this wasn’t true, Sen’s conclusions could not hold.

So, can we conclude the opposite? If a remediable social problem exists, can we say that it is because people don’t consider it a sign of failure?

13

Chris Bertram 01.06.05 at 9:46 am

DaF is a _very_ poorly edited and repetitive book. I leafed through without checking your page ref and, when the formula is repeated on p. 170 it concerns a functioning _multiparty_ democracy. Zimbabwe is again cited as an example, however ….

14

Chris Bertram 01.06.05 at 9:53 am

See also p. 183 where there is a contrast drawn between democracies and one-party states. Contra Daniel, therefore, I think it clear that having elections in which more than one party stands _is_ a necessary condition of a state being a democracy for Sen. One has to conclude, therefore, that Sen didn’t have a grip on the facts about Zimbabwe when he wrote as he did.

15

Motoko 01.06.05 at 9:53 am

Sam: It is the India-China comparison, post-1947/1949, that is perhaps most telling for Sen’s analysis…

Wasn’t one of Sen’s points that Maoist China’s record, bad as it is, is actually better than India’s, because over the years poverty in India resulted in more deaths from starvation and malnutrition than the Chinese famines?

16

John Isbell 01.06.05 at 12:04 pm

I once had dinner with the Kenyan education minister at Newark airport, and she remarked that the guy after Moi hadn’t even put his face on the money. And that Moi wasn’t that bad…

17

Donald Johnson 01.06.05 at 12:29 pm

Motoko makes the same point I made in an earlier thread–I never went back to see if anyone responded to it. It might be true that democracies don’t suffer famines (the Irish and the Indians under British rule didn’t have political rights, which is presumably why they aren’t a counterexample) but there is less to that statement than meets the eye, because most people who die from lack of food don’t do it in spectacular famines. As DD says, famines are singled out because lots of people die in a short time–if even larger numbers of people die in a more spread-out manner, the press and most pundits and even most of Sen’s alleged readers ignore it. Sen doesn’t ignore this—he points out that Maoist China actually has a better overall record in terms of increased life expectancy and fewer deaths from malnutrition than democratic India during the same period. But virtually all of the commentary on Sen ignores what he says about this and instead focuses on the nice warm fuzzy claim that democracies never suffer from famine. People either haven’t really read much of Sen (neither have I, frankly) or they are using his work in a propagandistic way. They might be afraid of sounding like apologists for Mao, but that’s silly, because you can say this about Maoist China while also acknowledging that Mao was one of the worst tyrants in history. As Sen also points out, the Indian state of Kerala achieved the same or better results compared to China and because they are a democracy, they did it without massive state violence, concentration camps, and spectacular famines.

18

Sam 01.06.05 at 1:46 pm

The key distinction here is malnutrition v. famine. Are they sufficiently different things? Sen’s point about democracy and famine is meant to explain very large numbers of deaths in a short amount of time, as opposed to deaths spread out over longer periods. If you accept the distinction, then, it seems to me, his argument has some merit. If you reject the distinction, you are not really dealing with his argument. Democracy may well have a role in preventing famines, but not in solving malnutirtion. But the question – and the title of the post – is famines. And, by the way, should we add Stalin’s famines to the list?

19

james 01.06.05 at 2:24 pm

Both Democratic India and Maoist China received outside assistance for food aid / food production during the time in question. This needs to be taken into account when comparing the two.

20

Mrs Tilton 01.06.05 at 2:33 pm

the Irish … under British rule didn’t have political rights

If you are referring to the Irish at the time of the famine, precisely what political rights did they not have? Presumably, you are thinking primarily about active and passive electoral rights, which were for a long time denied to UK Roman Catholics (not only those of them who were Irish).

First, it bears pointing out that ‘Irish’ != ‘catholic’. Second, catholics gained the ability to take a seat in Parliament 15 years before the famine began; they had gained the right to vote even earlier. (Admittedly, O’Connell’s deal on Emancipation did reduce the number of those eligible to vote by increasing the property qualification).

Post-emancipation, there remained a few miscellaneous disabilities that affected RCs. (They still do, in places that remain part of the UK.) These are not likely to have much impact, however, on a government’s willingness or ability to respond appropriately to a famine.

It was not a lack of political rights in the Irish that caused the government of the day to mismanage the famine as it did. It might have been a lack of political clout. A few decades after the famine, Irish parliamentarians had learned how to make governments notice them. Had that been the case at the time of the P. infestans outbreak, the results might have been less terrible.

21

Sebastian Holsclaw 01.06.05 at 3:24 pm

“Sebastian has a point – although not necessarily the one he thinks he’s making. :^)”

No, I agree that Sen may very well be using too expansive terms in the argument (actually too limited in that case). But I am absolutely sure that derek is.

22

Giles 01.06.05 at 3:38 pm

“I was originally just going to point out that the only African country which has managed to stay clear of famines entirely since independence is Kenya,”

what about South Africa?

23

abb1 01.06.05 at 4:21 pm

Expand “famine” into “poverty” and the USA looks less and less like a functioning democracy.

I think it’s perfectly reasonable to argue that high income/wealth inequality is an odd, unexpected characteristic of what one could reasonably identify as ‘functioning democracy’.

24

Donald Johnson 01.06.05 at 8:45 pm

To Mrs. Tilton–

Sounds like I was wrong. My grasp of Irish history was never that firm to begin with and I’ve probably forgotten half of the miniscule amount I once knew.

As for Sen and democracy and famines and malnutrition, I wasn’t arguing with Sen–from what I’ve read of him, he seems more nuanced than the people who constantly cite him on democratic India vs. Maoist China without ever mentioning the malnutrition deaths.

Incidentally, the book he wrote with Dreze on this (Hunger and Public Action, I think) used to be online, but last I looked it wasn’t anymore, probably because someone reprinted it.

25

Mrs Tilton 01.07.05 at 12:13 am

Och, Donald a mhic, don’t trouble yourself. If the Irish themselves had a better (or worse) grasp of Irish history, the world might be a better place for it.

Anyway, on the big picture I tend to agree with you. As I began reading Daniel’s post I was prepared to be all indignant. After reading the whole thing, I have the impression that what he is saying is that people who want to bandy Sen On Famines And Democracy about ought first to go read Sen On Famines And Democracy. And I can hardly quarrel with that.

26

Shai 01.07.05 at 12:33 am

“Since the word “democracy” has been abused …”

I remember from a comparative politics course a couple of years ago that there are several indexes for democracy, freedom whatever. But I’m a little too lazy to look up the specific reference in my notes. Thanks for nothing you say.

27

Conrad Barwa 01.07.05 at 1:22 pm

I’ve never been a big fan of Sen’s dictum that “democracies don’t have famines” – I’ve always regarded it as being a slogan on a par with “no two countries which have a McDonalds have ever gone to war with each other”.

I think this is somewhat of a later development; Sen’s original and really pioneering work was to show not that democracies somehow never have famines but to debunk the old FAD thesis on what causes famines – in his seminal 1981 book which takes an in-depth look at several famines and provides a formal theory of what causes them this is explored and explained at length. What is really remarkable is that by going over the micro-level statistics, Sen shows how regions and states that were meant to be suffering from famines did not see any collapse in per capita food availability and were actually net food exporters during famine periods. The association with democracy tends to be a later one and is specific to outbreaks of real droughts that lead to famines (as opposed to political causes or shifts in relative prices of commodities) and here he is correct to point out that a democratic state that is relatively stable and has a free media will not be able to survive non-action unlike a totalitarian state. India and China are the two examples that are usually cited in this regard, of course, however as pointed out; the real discontinuity holds better wrt to other factors such as fertility ratios and female literacy and that two for only specific regions within the respective countries. Perhaps a better way to phrase his dictum, which it must be said he never really puts in stark terms, is that no government in a functioning democracy could survive if it didn’t effectively try and combat a famine that occurred. More importantly, no government in such a state could ever get away with actually engineering a man-made famine (why any such government would want to do so is difficult to understand, on a national level at least).

It is this “plannability” of famine mitigation which is at the heart of Sen’s argument about famines and democracy. The actual reason why “democracies don’t have famines”, according to Sen, is as simple as pie. Think about it this way; it would be a very bad famine indeed that killed even as much as 10% of the population. The 10% of your population who are vulnerable to a famine are usually the poorest 10%, so their consumption is often as little as 3% of your GDP. And the famine comes about when their consumption drops below subsistence level, not when it disappears entirely. So basically, it is reasonable to suppose that the famine-vulnerable population could be brought back to their original level of consumption by diverting as little as 1.5% of GDP, and a government which can’t divert one and a half per cent of a country’s output is not a government worthy of the name.

Actually, no this is not correct. Even better than Griffith’s book, to understand how aid and redistribution can go wrong in LDC contexts is P. Sainath’s book “Everybody Loves a Good Drought” which looks at how in certain very well known cases of extreme poverty or hunger, most notriously Kalahandi, this supposed democratic link between state effectiveness and famine relief breaks down. Kalahandi district in Orissa is a prime example of a micro-level failure of Sen’s theory in that recurrent famine conditions have happened leading to debasing incidents such as debt-bondage and sale of children, which were picked up by the national media and then caused an uproar, leading to massive political attention including several prime ministers flying out to the area but which had little if any positive net impact. It was Rajiv Gandhi, I think who said that only about 10%-15% of the money the state distributes towards the rural poor reaches them and this was in the mid-80s when things were better; so to be able to distribute 1.5% of GDP and have it reach the intended beneficiaries will require a much larger infusion of money or a revolution of governance. Most govts like India can divert 1.5% of GDP, the problem is that they can’t make sure it reaches the intended destination; this might make them unworthy of the name but then this is perhaps not what they were configured to do. The Indian state these days likes to see itself as a nascent global power and aspirant to a permanent UNSC seat; it still can’t however prevent several thousand children from starving to death in one of its richest states.

28

Dan Hardie 01.07.05 at 4:58 pm

Conrad, thanks for the reference, and I’ll certainly try to read it. But to pick up on one point:

‘It was Rajiv Gandhi, I think who said that only about 10%-15% of the money the state distributes towards the rural poor reaches them.’
I really don’t trust this. Firstly I’m no expert on Indian politics, but Rajiv Gandhi always sounded like a shyster of the first order.

Secondly, this level of failure may well apply to development aid, but with famine relief we’re not talking about development aid. You need to supply a few skilled personnel plus bulk quantities of basic foodstuffs, potable water, medicines and public health (ie clean toilets in refugee camps), and, perhaps, a few seeds or items of livestock to persuade rural refugees back to their lands. There is much less opportunity for graft or waste in bulk-buying those items than there is in running many ‘development’ projects.

But thirdly, what you’ve said is a very valid criticism, or perhaps enlargement, of Sen’s work: the Indian sociologist Amrita Rangasami made the same points in 1981 in Delhi’s ‘Economic and Political Weekly’. Hunger among the rural poor, as Sen notes, only rarely reaches the point of ‘famine’- ie mass mortality. Sen wrote about ‘coping mechanisms’ and Rangasami pointed out that these include not just such things as reducing levels of food consumption, begging food from religious organisations or richer neighbours etc but also moving off land, becoming indentured labourers etc.

I agree that the real merit of Sen’s work is in his conclusive demolition of the Food Aggregate Deficiency thesis (ie the ‘common sense’ position that ‘people starve because there isn’t enough food nearby’). He’d accomplished this by 1981, and thereafter he started looking for a more positive account of what does cause famine, rather than what doesn’t. One of the reasons that it’s so hard to pin down his thinking on this is that he kept writing new books and articles (often with Jean Dreze)- and, to his credit, changing his mind when challenged. In the mid-80s, he did seem to say ‘democracies don’t suffer famines’; then it was ‘a free press will always prevent famines’; and Dsquared’s post is the best exposition I’ve read of the position he’s finally arrived at.

Umpteenth time I’ve recommended this, but Alex De Waal’s ‘Famine Crimes’ contains an excellent discussion, and history, of at least some of the problems associated with food aid to the starving. As De Waal notes, Sen’s demolition of the FAD was at least implicit in the Famine Codes of the British Raj in India, and was certainly anticipated by some of the pre-Raj Hindu writers of books on governance.

Btw, another interesting point:on the definition of famine as ‘an episode of mass deaths due to starvation rather than malnutrition’: De Waal’s ‘Famine that Kills’, a demographic study of the causes of death in a Sudanese famine in the early ’80s, reached the conclusion that the cause of mortality in the overwhelming majority of cases in the famine he studied was not malnutrition but diarrhoea due to lack of potable water and/or epidemics due to the lack of sanitation among refugees on the move. What killed people, in other words, was not the lack of food but the fact that the lack of food caused them to become refugees and thus exposed them to other hazards. No doubt this is untrue of famine afflicting, say, the poor in an urban context (eg Jews locked up in the Kiev ghetto in the Second World War, refugees in an African city today) but De Waal concluded that it was poor water and epidemics that killed in most famines, not simple lack of calories.

29

Dan Hardie 01.07.05 at 4:59 pm

Conrad, thanks for the reference, and I’ll certainly try to read it. But to pick up on one point:

‘It was Rajiv Gandhi, I think who said that only about 10%-15% of the money the state distributes towards the rural poor reaches them.’
I really don’t trust this. Firstly I’m no expert on Indian politics, but Rajiv Gandhi always sounded like a shyster of the first order.

Secondly, this level of failure may well apply to development aid, but with famine relief we’re not talking about development aid. You need to supply a few skilled personnel plus bulk quantities of basic foodstuffs, potable water, medicines and public health (ie clean toilets in refugee camps), and, perhaps, a few seeds or items of livestock to persuade rural refugees back to their lands. There is much less opportunity for graft or waste in bulk-buying those items than there is in running many ‘development’ projects.

But thirdly, what you’ve said is a very valid criticism, or perhaps enlargement, of Sen’s work: the Indian sociologist Amrita Rangasami made the same points in 1981 in Delhi’s ‘Economic and Political Weekly’. Hunger among the rural poor, as Sen notes, only rarely reaches the point of ‘famine’- ie mass mortality. Sen wrote about ‘coping mechanisms’ and Rangasami pointed out that these include not just such things as reducing levels of food consumption, begging food from religious organisations or richer neighbours etc but also moving off land, becoming indentured labourers etc.

I agree that the real merit of Sen’s work is in his conclusive demolition of the Food Aggregate Deficiency thesis (ie the ‘common sense’ position that ‘people starve because there isn’t enough food nearby’). He’d accomplished this by 1981, and thereafter he started looking for a more positive account of what does cause famine, rather than what doesn’t. One of the reasons that it’s so hard to pin down his thinking on this is that he kept writing new books and articles (often with Jean Dreze)- and, to his credit, changing his mind when challenged. In the mid-80s, he did seem to say ‘democracies don’t suffer famines’; then it was ‘a free press will always prevent famines’; and Dsquared’s post is the best exposition I’ve read of the position he’s finally arrived at.

Umpteenth time I’ve recommended this, but Alex De Waal’s ‘Famine Crimes’ contains an excellent discussion, and history, of at least some of the problems associated with food aid to the starving. As De Waal notes, Sen’s demolition of the FAD was at least implicit in the Famine Codes of the British Raj in India, and was certainly anticipated by some of the pre-Raj Hindu writers of books on governance.

Btw, another interesting point:on the definition of famine as ‘an episode of mass deaths due to starvation rather than malnutrition’: De Waal’s ‘Famine that Kills’, a demographic study of the causes of death in a Sudanese famine in the early ’80s, reached the conclusion that the cause of mortality in the overwhelming majority of cases in the famine he studied was not malnutrition but diarrhoea due to lack of potable water and/or epidemics due to the lack of sanitation among refugees on the move. What killed people, in other words, was not the lack of food but the fact that the lack of food caused them to become refugees and thus exposed them to other hazards. No doubt this is untrue of famine afflicting, say, the poor in an urban context (eg Jews locked up in the Kiev ghetto in the Second World War, refugees in an African city today) but De Waal concluded that it was poor water and epidemics that killed in most famines, not simple lack of calories.

30

dsquared 01.07.05 at 10:13 pm

I think Dan has to be right and Rajiv Gandhi wrong on the subject of disaster aid. This is why the army are always so prominent in disaster relief efforts; as organisations, armies are fantastic at delivering things to people (after all, their main function in life is the delivery of small bits of metal to people who don’t want them).

31

Conrad Barwa 01.08.05 at 2:43 am

Dan,

I really don’t trust this. Firstly I’m no expert on Indian politics, but Rajiv Gandhi always sounded like a shyster of the first order.

Well, I don’t want to get sidetracked into a discussion about the integrity of Indian PMs, but Rajiv Gandhi wasn’t all that bad, wasn’t all that great either. He needs to be seen in a context where pretty much every single PM that has led a govt which has lasted at the centre since the mid-60s have been unscrupolous manipulators of the first order. I think my point wasn’t to comment on Rajiv Gandhi record as a PM but just to point out how ‘leaky’ and difficult to manage state efforts in this area was and indeed is.

Secondly, this level of failure may well apply to development aid, but with famine relief we’re not talking about development aid. You need to supply a few skilled personnel plus bulk quantities of basic foodstuffs, potable water, medicines and public health (ie clean toilets in refugee camps), and, perhaps, a few seeds or items of livestock to persuade rural refugees back to their lands. There is much less opportunity for graft or waste in bulk-buying those items than there is in running many ‘development’ projects.

I am afraid not; one would this should be the case but it isn’t. I know this personally and professionally; the lead IAS officer who was in charge of relief efforts for the Latur earthquake victims was actually tranferred out in what was a well known case at the time because he was refusing to allow local legislators and party fixers to dip their hand in the funds. A more recent example is the general mis-handling of rehabilitation efforts for the Orissa cyclone that occurred a couple of years ago; the only effective work was carried out by volunteers from Andhra Pradesh sent by the Naidu govt which led to many Oriya beneficiaries adopting pro-TDP armbands and slogans as a mark of acknowledgement and gratitude. I woulde reverse your reasoning and say that there is even more scope graft and waste since emergency releif often means that the aid has to reach the targets in time with the minimum of delay or obstruction – which simply put can be very costly in human terms. In a situation analogous to wartime, this can allow very large exploitative rents to be extracted by those in a position to do so.

In the mid-80s, he did seem to say ‘democracies don’t suffer famines’; then it was ‘a free press will always prevent famines

He sort of moved around it a bit; there is the volume Hunger and Public Action that he wrote with Jean Dreze that looks at the specific success of the Maharastra state govt in dealing with famine and the edited collection of papers in ‘The Political Economy of Hunger’ has a few chapters on the role of media and other institutions. Bob Currie has an excellent book on Kalandi, ‘Poverty and Hunger’ that actually shows some of the weaknesses in Sen’s logic. Localised famines led to a war of words being conducted in the media whereby the state govt actually took out multi-page adverts insisting that no famine was taking place, and in many districts clear provisions for action in the case of malnutrition deaths were simply ignored by attributing death to other causes such as eating of mahua fruits and other inedible materials – circumventing the declaration of mass hunger conditions. Of course, the reason why people were eating tree bark and other such stuff was the fact that they simply had run out of edible food to eat. All this evades the real paradox of persistent hunger in democracies and this is how such a large proportion of the population can go unfed while food is literally rotting in grain godowns or being exported at subsidised prices; which was the case for the last 5-7 years in India.

Umpteenth time I’ve recommended this, but Alex De Waal’s ‘Famine Crimes’ contains an excellent discussion, and history, of at least some of the problems associated with food aid to the starving. As De Waal notes, Sen’s demolition of the FAD was at least implicit in the Famine Codes of the British Raj in India, and was certainly anticipated by some of the pre-Raj Hindu writers of books on governance.

Yeah, De Waal is a good guy; I would also recommend his most recent paper on ‘New Variant Famine’ in the Lancet which looks at the link between AIDs and malnutrition. Two pre-colonial thinkers that preceded Sen in this regard would be MG Ranade and Dadabhai Naroji; both influential in what could be termed as the Bombay School of political economy.

No doubt this is untrue of famine afflicting, say, the poor in an urban context (eg Jews locked up in the Kiev ghetto in the Second World War, refugees in an African city today) but De Waal concluded that it was poor water and epidemics that killed in most famines, not simple lack of calories.

Sure, I think this would be true of many famines. Actually dying of hunger is a very long and drawn out process, most people would be finished off by disease or epidemics long before this. As I mentioned earlier many try to eat things that are either inedible or poisonous and die of food poisoning – conveniently allowing the govt to pass the buck on there actually being an outbreak of malnutrition deaths.

Dsquared,

I think Dan has to be right and Rajiv Gandhi wrong on the subject of disaster aid. This is why the army are always so prominent in disaster relief efforts; as organisations, armies are fantastic at delivering things to people (after all, their main function in life is the delivery of small bits of metal to people who don’t want them).

Yes and no. The army isn’t always called out in such instances and even when it is, they are not given authority over the procurement and strategic allocation of aid; I remember that the first active mission I underwent was in assisstance to village communities flooded in Punjab; but we were always under civilian control as far as actual distribution of aid went – ultimately we could only do a good job if the civil authorites didn’t mess around with the disbursement of supplies. The army does play a stronger role in preventing any breakdown of law and order and of course, in actually physically saving people or transporting them from hazardous areas to safe ones. As is so often the case, the military is mainly a tool and they can be used efficiently and effectively or poorly and inappropriately; this is primarily a political question not a tactical or operational one. Even when we have to ‘deliver small bits of metal to people who don’t want them’ we don’t always get to choose the who, why or where; which is often very important.

32

Conrad Barwa 01.08.05 at 2:44 am

Dan,

I really don’t trust this. Firstly I’m no expert on Indian politics, but Rajiv Gandhi always sounded like a shyster of the first order.

Well, I don’t want to get sidetracked into a discussion about the integrity of Indian PMs, but Rajiv Gandhi wasn’t all that bad, wasn’t all that great either. He needs to be seen in a context where pretty much every single PM that has led a govt which has lasted at the centre since the mid-60s have been unscrupolous manipulators of the first order. I think my point wasn’t to comment on Rajiv Gandhi record as a PM but just to point out how ‘leaky’ and difficult to manage state efforts in this area was and indeed is.

Secondly, this level of failure may well apply to development aid, but with famine relief we’re not talking about development aid. You need to supply a few skilled personnel plus bulk quantities of basic foodstuffs, potable water, medicines and public health (ie clean toilets in refugee camps), and, perhaps, a few seeds or items of livestock to persuade rural refugees back to their lands. There is much less opportunity for graft or waste in bulk-buying those items than there is in running many ‘development’ projects.

I am afraid not; one would this should be the case but it isn’t. I know this personally and professionally; the lead IAS officer who was in charge of relief efforts for the Latur earthquake victims was actually tranferred out in what was a well known case at the time because he was refusing to allow local legislators and party fixers to dip their hand in the funds. A more recent example is the general mis-handling of rehabilitation efforts for the Orissa cyclone that occurred a couple of years ago; the only effective work was carried out by volunteers from Andhra Pradesh sent by the Naidu govt which led to many Oriya beneficiaries adopting pro-TDP armbands and slogans as a mark of acknowledgement and gratitude. I woulde reverse your reasoning and say that there is even more scope graft and waste since emergency releif often means that the aid has to reach the targets in time with the minimum of delay or obstruction – which simply put can be very costly in human terms. In a situation analogous to wartime, this can allow very large exploitative rents to be extracted by those in a position to do so.

In the mid-80s, he did seem to say ‘democracies don’t suffer famines’; then it was ‘a free press will always prevent famines

He sort of moved around it a bit; there is the volume Hunger and Public Action that he wrote with Jean Dreze that looks at the specific success of the Maharastra state govt in dealing with famine and the edited collection of papers in ‘The Political Economy of Hunger’ has a few chapters on the role of media and other institutions. Bob Currie has an excellent book on Kalandi, ‘Poverty and Hunger’ that actually shows some of the weaknesses in Sen’s logic. Localised famines led to a war of words being conducted in the media whereby the state govt actually took out multi-page adverts insisting that no famine was taking place, and in many districts clear provisions for action in the case of malnutrition deaths were simply ignored by attributing death to other causes such as eating of mahua fruits and other inedible materials – circumventing the declaration of mass hunger conditions. Of course, the reason why people were eating tree bark and other such stuff was the fact that they simply had run out of edible food to eat. All this evades the real paradox of persistent hunger in democracies and this is how such a large proportion of the population can go unfed while food is literally rotting in grain godowns or being exported at subsidised prices; which was the case for the last 5-7 years in India.

Umpteenth time I’ve recommended this, but Alex De Waal’s ‘Famine Crimes’ contains an excellent discussion, and history, of at least some of the problems associated with food aid to the starving. As De Waal notes, Sen’s demolition of the FAD was at least implicit in the Famine Codes of the British Raj in India, and was certainly anticipated by some of the pre-Raj Hindu writers of books on governance.

Yeah, De Waal is a good guy; I would also recommend his most recent paper on ‘New Variant Famine’ in the Lancet which looks at the link between AIDs and malnutrition. Two pre-colonial thinkers that preceded Sen in this regard would be MG Ranade and Dadabhai Naroji; both influential in what could be termed as the Bombay School of political economy.

No doubt this is untrue of famine afflicting, say, the poor in an urban context (eg Jews locked up in the Kiev ghetto in the Second World War, refugees in an African city today) but De Waal concluded that it was poor water and epidemics that killed in most famines, not simple lack of calories.

Sure, I think this would be true of many famines. Actually dying of hunger is a very long and drawn out process, most people would be finished off by disease or epidemics long before this. As I mentioned earlier many try to eat things that are either inedible or poisonous and die of food poisoning – conveniently allowing the govt to pass the buck on there actually being an outbreak of malnutrition deaths.

Dsquared,

I think Dan has to be right and Rajiv Gandhi wrong on the subject of disaster aid. This is why the army are always so prominent in disaster relief efforts; as organisations, armies are fantastic at delivering things to people (after all, their main function in life is the delivery of small bits of metal to people who don’t want them).

Yes and no. The army isn’t always called out in such instances and even when it is, they are not given authority over the procurement and strategic allocation of aid; I remember that the first active mission I underwent was in assisstance to village communities flooded in Punjab; but we were always under civilian control as far as actual distribution of aid went – ultimately we could only do a good job if the civil authorites didn’t mess around with the disbursement of supplies. The army does play a stronger role in preventing any breakdown of law and order and of course, in actually physically saving people or transporting them from hazardous areas to safe ones. As is so often the case, the military is mainly a tool and they can be used efficiently and effectively or poorly and inappropriately; this is primarily a political question not a tactical or operational one. Even when we have to ‘deliver small bits of metal to people who don’t want them’ we don’t always get to choose the who, why or where; which is often very important.

33

Dan Hardie 01.08.05 at 3:34 pm

Conrad- this is fascinating. Let me know if you ever start a blog. You’re arguing from experience and I’m arguing from my reading, so I think you win.

Would you quibble with the conclusion (reached by both Sen and De Waal) that post-Independence Indian governments have effectively abolished famine? You’ve come up with a number of counter-examples which make me suspect you don’t quite agree with this.

34

Dan Hardie 01.08.05 at 3:43 pm

‘I woulde reverse your reasoning and say that there is even more scope graft and waste since emergency releif often means that the aid has to reach the targets in time with the minimum of delay or obstruction – which simply put can be very costly in human terms. In a situation analogous to wartime, this can allow very large exploitative rents to be extracted by those in a position to do so.’

I don’t doubt your word, but how is it that this occurs in modern-day India? You’ve got a free press, Parliamentary and State governments with the power to hold enquiries, a judiciary modelled on the UK’s independent system: how is it that people don’t avoid graft on famine contracts because they’re afraid of whistleblowing and subsequent criminal prosecution?

35

s 01.09.05 at 12:40 am

This also makes me think of China’s government today. Their emphasis on growth in quantities (they foster competition in price) is driven by a fear that, if the economy slows down, social discontent will endanger the government’s stability. It isn’t really that the government cares very much about the welfare of the people, I think, but rather that they know that if hunger is widespread, they are in trouble.

36

Jonathan Edelstein 01.10.05 at 1:32 am

I think it clear that having elections in which more than one party stands is a necessary condition of a state being a democracy for Sen. One has to conclude, therefore, that Sen didn’t have a grip on the facts about Zimbabwe when he wrote as he did.

Zimbabwe wasn’t a de jure one-party state in 1999. It was a one-party state de facto – the opposition held three of 150 parliamentary seats – but then again, so was Botswana. The factors that made Zimbabwe non-democratic in 1999 had more to do with crony capitalism, the influence of the security forces and the subornation of the courts than the absence of opposition parties.

37

dsquared 01.11.05 at 5:16 pm

But Zimbabwe was a one-party state between 1988 (merger of ZANU and ZAPU) and 1999 (formation of MDC). That’s the period during which Sen was doing a lot of the work that ended up in DaF.

Comments on this entry are closed.