You are absolutely right that supply conditions play a role, but perhaps that makes it all the more imperative that policies directed towards increasing the money incomes of the poor should also consider the simultaneous need for strategies to improve supply.
In Iran the sanctions are not new, though it is true that the cumulative impact would lead to progressively constrained availability of many essential commodities. But then that has to be factored in when seeking to substitute in-kind transfers with cash transfers. What the Iranian government currently seems to be trying to do is cope with the price rationing that is implicit in higher prices of basic items by providing a quantity rationing of a certain minimum amount of basic items per households, similar to what was provided for example during the war years. But now the public distribution channels are missing or have been destroyed, so it is practically much more difficult.
In India, to take a different example, the employment guarantee programme MNREGA has been blamed (even by the Prime Minister!) for increasing demand for essential food items and thereby fuelling food inflation. While the debate is still on about that (there is a much more significant role of cost push inflation driven by rising fuels costs which obviously enter into all other prices) it is obvious in any case that a strategy like MNREGA should be accompanied by policies that enable expanded production of food and other essential items, including by making small holder cultivation more viable which may in turn require some subsidies that encourage more efficient and sustainable agricultural production.
So rather than having a knee-jerk reaction about subsidies being bad (including fuel subsidies) it makes more sense to assess the specific context and goals of particular subsidies and see whether these can be usefully achieved in other ways. To that extent, counterposing fuel subsidies with direct income transfers to the poor may not be so useful, since the goals of fuel subsidies in particular contexts need not be "poverty alleviation" per se, but may be about keeping inflationary pressure down in the economy or encouraging certain types of production, etc. So maybe these can be refined so that they are better able to meet those goals and reduce the "leakages" to undesirable purposes.
Incidentally, I am not against cash transfers as such, in fact I am all for them when they are additions to public provision of essential goods and services to enable universal access, rather than substitutes for such provision. But even with such additional cash transfers (pensions, child support, scholarships, etc.) the potential inflationary impact needs to be borne in mind and appropriate policies formulated accordingly.
Jayati GhoshProfessor of economics at Jawaharlal Nehru University, New Delhi, and Executive Secretary of International Development Economics Associates (Ideas)