The Moorish Wanderer

Mean-Tested Compensation Fund and Regional differences

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on June 18, 2012

I have reached a certain level of my graduate life where I need to sort out my writings: less of free-style blog-posts, and more serious, academic papers. I guess from now on I might be less inclined to argue forcefully on economic issues, and the reader can make up their mind as to the need to present such and such argument for such and such claim.

Last post was about a simple and relatively inexpensive scheme designed to induce a mechanism design such that only those households below the median income to genuinely benefit from the subsidy, so as to avoid an excessive compensation whose actual beneficiaries are those at the top. The idea is to look at the median consumption, subsidize it, and provide a commensurate cash relief to any household claiming it. Hopefully, the wealthiest would back away from the scheme, since that would mean a drastic reduction in their consumption habits, and each would pay the true relative price of their consumption bundle.

This post deals with the finer details of such program; the idea is to identify regional means of consumption, hopefully their median consumption, and from then on apply the same method nationwide; the argument behind the region-based discrimination is obviously is its cost-efficiency: a nationwide median understates the discrepancies between regions, just as a mean does – you might understand it as an average median, which brings equal problems in terms of mechanism design strategy-proofness. And based on the figures provided by HCP, we do observe a great deal of discrepancies, which signals to a lower cost for cash-relief.

According to HCP’s “Comptes régionaux PIB régional et dépenses de consommation afinale des ménages“, Grand Casablanca and Rabat regions account for 35% of total GDP, but only 26.4%. The 9% discrepancy captures what might be construed as a typical illustration of Keynes’ consumption function: richer households tend to consume less relative to their output, and they are likely to devote less of their new income when it grows (the so-called marginal propensity of consumption) But still, two regions out of 16 account for the fifth of household consumption in Morocco means a reasonable claim can be made as to the distribution of actual recipient of the Compensation Fund subsidies. As a matter of fact, the subsidies goes to the wealthiest households in metropolitan areas, whose relative consumption to household is significantly lower compared to nationwide figures; in absolute terms though, the average Casawi household consumes twice as much as their opposite number in Taza or Alhuceimas.

Regional distribution of actual subsidy provided by the median-based cash relief scheme. Poorer regions tend to benefit above the 20% nationwide relief. (big up for Kingstoune and the neat SVG master map)

The median regional household consumption is in the neighbour of 93,760 dirhams per annum, with extreme values as high as 139,810 in Rabat-Salé-Zemmour and as low as 72,460 dirhams per household in the Gharb.

The scheme would provide cash relief on the basis of the regional median household, with those below the median benefiting up to 2,700 dirhams in excess, precisely because they can claim higher levels than their actual consumption, a significant enhancement for their purchasing power.

On the other hand, those above the median would have to sacrifice more than 25,000 dirhams of consumption to be eligible for that cash-relief scheme, something that represents a net loss of 21% of their current consumption. The arbitrage left to the upper households is pretty straight forward: either accept to lose 21% of their current consumption immediately, or accept to pay a higher price gradually.

On the basis of a 60% population eligible for the cash-relief system, the net cost for the Compensation Fund would be valued at around 25,5Bn dirhams, 400 Million dirhams less than the nationwide-based median consumption bundle. This is the best evidence yet that local standards are best in determining the weighted-average median consumption bundle on the basis of which households are eligible for the cash relief scheme. These figures are based on 2011 estimates, which explains the discrepancies with respect to the figures put up in the latest post, since these are 2010 estimates. This means that the cost of the Subsidy fund can be maintained constant in real terms, i.e. relative to inflation, but also with respect to the nationwide GNI growth.


Food and Consumption Subsidy: Below the Median

Posted in Dismal Economics, Flash News, Moroccan Politics & Economics, Morocco by Zouhair ABH on June 9, 2012

I did not watch our Head of Government’s performance on TV. But I guess by now, the Moroccan public will take a keen interest in the essential mechanism of economic interactions. Mr Benkirane’s favourite catchphrase “God’s Will” might not be as helpful as he makes out.

A strong negative correlation between Consumption and Net Exports, compounded by the strong contribution of the former to GDP growth

First off, the graph opposite just shows the insanity and how unsustainable the subsidies are: Moroccan has a structural deficit in its trade balance, and the compensation fund does not subsidize household consumption, but chiefly subsidizing the trade deficit. As such, the Moroccan government’s hands are tied: if they decide to reform radically the compensation fund, or address the problem of trade balance and balance of payments, this would mean doing away with a reliable source of growth -household consumption- and a lucrative source of fiscal receipts.

Household consumption is a curse no longer in disguise: the weaker foreign demand for Moroccan exports from the Eurozone, the stronger our economy will have to rely on domestic consumption, the heavier the weight of subsidies on public finances, with a paradoxical effect on government receipts, but only up to a point.

The idea is thus to achieve a three-fold objective:

First, reform the compensation fund in an effort to address the structural weaknesses of the Budget.

Second, direct relief to households genuinely in need for it.

Third, balance domestic consumption so as to make Exports a viable substitution in terms of growth contribution.

The Iranian example provided by fellow blogger Omar El Hayani is too inflationary, and I am afraid his computations were a bit far-off base: suffice to say these computations should be done at the household level (400 dirhams per person allowance does not sound to be a viable program at any rate). A household-based direct subsidy in form of cash relief ought to perform better, with a nation-wide benchmark.

The idea is simple enough to avoid the ritual pitfalls of ‘Institutional Shortcomings’ (codeword for government corruption) by applying a small device from Game Theory properties: the subsidy is computed on the basis of a composite basket of goods a median household usually spends money on. The advantage of such a mechanism is that all those households below the Median would automatically benefit from it, regardless of their declared type of consumption. As for the better off households above the median, they have the choice between keeping on with their existing patterns of consumption -and charged more for it- or reduce their consumption absolute levels so as to match the subsidized median basket eligible for the cash relief, which results in a reduction in their consumption in absolute value. This is a win-win outcome: poorer households observe their purchasing power is stable or improving, and the wealthier Moroccans are given the opportunity to pay the true price of their consumption.

This model has the advantage of deflecting inflation away from the vulnerable households: lest we forget, about a third of household consumption in Morocco is concentred in the hands of 10% wealthiest households, these are the ones benefiting from the current system of subsidies, and these are the ones behind any sizeable inflationary shock.

Consider HCP’s households survey in 2000-2001: The average share of income devoted to food consumption established itself around 41%. The figure itself, when compared to the median share of consumption (47%) shows how skewed household consumption is in favour of the wealthiest, even though these consume only a little more than 30% of their disposable income.

Per these findings, the benchmark a household consist of 6 individuals, most likely to be 4 over time, with an annual gross income of 76,940 dirhams (2010 estimates).

Their annual consumption establishes itself around 36,100 dirhams per annum. If these are subsidized at 20%, the total cost for the Compensation Fund would be established around 25.89 Bn dirhams per year, in real terms. The median household is subsidized at 20%, the poorer 10% at around 42% of their consumption, and the real subsidy percentage would be closer to 25%, which only confirms how this particular scheme helps those who really need it.

Subsidized Pass the parcel: Compensation Fund

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco by Zouhair ABH on February 12, 2012

As per the latest Treasury survey figures, the Compensation Fund stands at about 41Bn – although the initial 2012 Budget bill provides for 45Bn; that represents 5.7% of GDP, 14% of total budget. It also represents 80% of total public investments for 2012. And finally, it seems all public subsidies equate total expenditure en education. And yet every annual budget statement pushes for a Compensation Fund reform:

L’accent sera mis, notamment, sur des questions liées à la réforme fiscale, à la masse salariale, à la réforme de la compensation… pour s’assurer de la soutenabilité des finances publiques à la lumière des exigences de développement économique et social auquel le Maroc est en droit d’aspirer.

But is the compensation fund worth the money? After all, the total resources allocated to the most important expense – or shall we say the most politically sensitive- remains wheat and sugar; According to the addendum on special funds to the Budget bill (Comptes Spéciaux du Trésor) the designated “Fonds de Soutien des Prix de certains Produits Alimentaires” is allocated with some 880 Million dirhams – and even that amount of money can be halved; either by targeting those subsidies (which hasn’t really been the case ever since subsidies were enforced) or by by setting up a task force within the Exchange Office or the Finance Ministry to trade in those commodities the Moroccan authorities consider to be valuable.

I’m looking at some of the ministry’s own figures regarding the observed average traded prices of sugar on international markets,  and they list traded average prices:

Les cours moyen du sucre brut et du blé tendre au titre du premier semestre 2011 s’élèvent respectivement à près de 652 $/t et 352 $/t contre respectivement 444 $/t et 172 $/t au titre de la même période de l’année 2010.

It is a bit strange the Sugar Index prices do not match MINEFI figures. Or perhaps looking for the cheaper alternative isn't a priority down there. (Bloomberg)

But it seems to me this is not true; not for sugar anyway. I have looked up two indexes for sugar, and the maximal value did not top up $ 360/ton – March 2011, if our government was wise enough to cover imports with futures. As for wheat, it seems the market price did not go beyond $350/ton since 2008; that  $ 20 doesn’t make a difference, and for the specific category of wheat, total cost to import was about MAD 6 Bn for 2010; that means each household had to shoulder 896 dirhams of direct expenses incurred from wheat imports. Total cost for imported sugar (of all kinds) would have been 3.3Bn, that is 511 dirhams per household

896 dirhams of basic cost computed from raw imported wheat; assuming negligible additional costs, and based on the regulated price of one loaf of bread, that means one household would have consumed an annual set of 690 loafs of bread; same computations for sugar based on an average price of 6 dirhams/kg would yield a theoretical consumption of 85kg per household. Gee-whiz statistics perhaps, but it paints a picture; international prices have only limited impact on purchasing power; as far as I can tell, domestic interactions account for a lot more, perhaps because of the strait-jacketing of specific prices and oligopolistic markets for goods such as wheat storage, sugar and distribution.

Now 2010 yielded an average annual income per household of about MAD 106,751, and based on HCP 2006/2007 household consumption survey, consume 41.3%, thus spending MAD 1812 in sugar and MAD 7847 in wheat (bread and otherwise) That’s a lot more than the raw cost for these imported goods.

This comparison suffers from major flaws however: it does not take into account other costs factored in the final product, and second, computing expenditure in average term tends to produce bias; indeed, compensation subsidies are not mean-tested; furthermore, figures are not that reassuring when one looks into the median expenditure, let alone the fact that average household does not take into account its intrinsic size. Median annual income has been MAD 77,000 and thus consume MAD 5288 in wheat and MAD 1244 in sugar per annum.

But the argument is basically sound: raw cost per household is too low to justify what might come to be a MAD 7,800 subsidy per household. Actually, that argument is strengthened by the disparities in consumption distribution; in fact, because the top 10% households (that’s 2 Million individuals, give or take) capture a third of total household consumption, they get more than their theoretical 10% share in subsidies -since those are indiscriminate- in fact they get three times more than average, and they get to benefit from 3/4 of total subsidies – this is derived from the fact that subsidies are distributed uniformly across population deciles.

I would argue it is too high a price to pay for an arcane system that may have been working when income dispersion was not that high: with a widening income, wealth and consumption gaps within the economy, an indiscriminate subsidy on food and other essential goods tends to favour those who capture the largest proportion of household consumption, i.e. the top 10%, or as one would like to label them, those who do not need the subsidy.

Furthermore, the compensation also encompasses the following:

à la compensation des produits de base, en l’occurrence le sucre et la farine. Le montant réglé dans ce cadre par prélèvement sur le compte intitulé “Fonds de soutien des prix de certains produits alimentaires” se monte à 880 MDH, auquel s’ajoutent des dépenses de 14.987 MDH prises en charge par le budget général (Chapitre des Charges Communes), dont 2.357 MDH au titre de la compensation des denrées alimentaires de base et 12.630 MDH destinés à couvrir la charge de compensation des produits pétroliers.

So as far as identified compensation-related items go, the industrial subsidy is 4 times larger than the more important matter of subsidizing food and edible goods. A call has been made by government officials to maintain an industrial status-quo that might not, after all, benefit the domestic economy; what if the cost of subsidizing oil-intensive industries was actually much higher than any transitional loss to alternative industries? And how come oil-related imports benefit from a 12.6Bn subsidy when the immediate household consumption of Butane is only 15%? Unless Butane is sold for a symbolic price (which is not the case as long as I can recall the commercial price of one butane cylinders) that subsidy goes primarily into business oil consumption;

So total expenses for subsidizing food is at most 3.2Bn, that’s less than 1% of total budget and 13% of projected deficit – which is so far quite a sustainable subsidy and an expenses ceiling for future mean-tested programs for targeted subsidies. As for oil-related subsidies, the same observations can be made regarding income/consumption gaps: on households’ sides, who owns/drives cars? on business’ side, which industries are hungry for oil? A valid counter-argument can be made for the poorer household deciles; how will they cook and keep their family warm? Well, given the fact that these account for 10%, and since HCP reports the following:

En 2007, 54,6% des ménages marocains disposaient d’une cuisinière à gaz. Selon le milieu de résidence, ce taux passe de 42,9% dans les campagnes à 61,6% dans les villes. Selon la classe de dépense, ce taux passe de 67,4% pour les ménages les plus favorisés à 37,1% pour les moins nantis.

[…] Au niveau national, 20% des ménages disposent d’un chauffe-eau à gaz. Ce taux varie de 1,2% pour les 20% les moins aisés à 45,9% pour les 20% les plus riches. En milieu urbain, le taux d’équipement est 10 fois supérieur à celui enregistré en milieu rural (30,3% contre 3,0%).

Furthermore, the bottom 20% households are twice less likely than the top 20% to own a gaz-based stove, and only 1.2% of those households own a gaz-based heaters.

This is to say that an inflated compensation fund, in all its individual components, is not due to the international hike in prices – only 4Bn out of the 45 have a direct impact on the livelihoods of Moroccan households;A willing (and strong-willed) finance minister can gradually start reforming the fund by first setting up a team to supervise pricing on international markets -by using commodity futures, since Morocco’s own demand is unlikely to tip futures’ valuation one way or the other- and get prices down by as much as 6% per annum for wheat for instance. Second, there must be a way to mobilize the services of the Inland Revenues (Direction des Impôts) HCP and the whole Finances ministry to design mean-tested tax deductions and breaks, as well as cash relief schemes to support as many as 661,000 households, about 5.3Mln individuals who do not have the means to sustain themselves; so far, a 7,000 annual relief (that’s a 23% boost on average income of the bottom decile) would cost about 4.7Bn and still be a relatively low burden on the budget, plus subsidy would go to those it can actually help. Third, subsidies to businesses cannot go on indefinitely; first because it skews Morocco’s industrial structure, and would give an unfair advantage to what might be obsolete and rent-seeking industries; subsidized fuel only boost the value of a taxi license, when the money can be used to improve public transportation.

The thing is, there is a way to halve subsidies gradually, though it means taking on many powerful lobbies. The only essential condition for such a scheme to succeed is political courage. I doubt Mr Baraka or his sidekick Mr Azami Idrissi, or even Mr Benkirane have what it takes to take the fight where it is necessary: suppressing rent, promoting innovation to boost growth.