The Moorish Wanderer

“Regional Solidarity”: Bums and Workaholics

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on September 10, 2012

Ever wonder how much of your taxpayer’s money went to other regions? Of course, if you are from Casablanca, or Agadir, you are entitled to ask if you are from Rabat on the other hand, not so much. Unfortunately however, some budgetary constraints prevent the curious inquirer to get the raw numbers from our administration. And so, I endeavour to crunch these available numbers together to get some idea of how things are computed.

Average regional GDP per Capita in these super-regions is 21% higher than nationwide GDP per Capita.

First, I start with the standard national accounting identity: Y=C+G+I+NX. (Output = Consumption +Government Spending + Investment + Net Exports)

In fact, I can even assume that equality is simplified to Y=C+G+I since most of our exports are concentrated on two seaports at the most (Casablanca alone attracts 42% of total export/import shipping ) and use data from the MINEFI paper on regional contributions to GDP, as well as an HCP survey from 2007. It is without much surprise that 5 regions concentrate about 60% of total GDP (Casablanca, Rabat, Marrakech, Tangiers and Souss) and a little less than half of total population. We can also safely assume productivity per capita in these regions is significantly larger, paradoxically because their respective occupation level of active population would be lower.

Why would I need the national accounting identity to check which regions rely on government subsidies and transfers? Well, it is a matter of simple economics: a thriving region would not necessarily have a high regional GDP – Soussa Massa has a relatively low GDP per Capita, yet it is one of the richest regions in Morocco (4th richest not including Raba-Salé). What matters really is how their regional GDP is formed; a wealthy, productive region should produce its own consumption and pay relatively high taxes – or at least close to nationwide levels.

The following results are based on computations of aggregates per capita: there is a logical enough argument to be made that poorer regions might be over-populated; as it turned out, richer regions tend to have larger populations, they are however more productive, even more so, given the fact their active population is actually smaller, when compared to nationwide occupation rate of active population as well as those of the poorer regions. Per capita results take the demographics out of the equation, and even the odds somewhat.

The initial point made about wealthy regions stems from the standard national accounting equation: regional output is (roughly) consumed, taxes or invested. A good point can be made as to how local output matches local consumption, i.e. food and other goods consumed in one region are not necessarily made there; after all, sea-fish consumed in Marrakesh has to come from a coastal city, and Melons down South in Laayun need to come from another, cooler, watery place. Still and all, productive regions are able to produce enough output to buy them their consumption from other regions. Those too poor to afford anything will have to rely on government subsidies, or else reduce their consumption to subsistence levels. six regions emerge in this case: the Southern provinces, Tadla-Azilal and Taza-Alhuceimas. Their cumulative contribution to total GDP is less than 10%, and their average GDP per capita is roughly that of Souss-Massa.

Taxes and Government spending however are a different place; government money levied from or spent on a region stays there. Unfortunately, we do not have the exact amount of government spendings per region, though the other side of the equation is out there: there is evidence about how much each region contributes to total fiscal receipts; As it turns out, the 5 super-regions contribute about 91.5% of 2011 fiscal receipts, about 138.2Bn that is. So the initial body of evidence is there: the richest regions tend to pay more taxes than they produce output, and if Rabat-Salé is excluded from computations, the 4 super-regions account for 74% of fiscal receipts, versus a little less than half of total GDP. In simple arithmetic, every 100 dirhams these 4 regions paid 19.1 of it in taxes, and these were transferred to other regions.

What is the difference between the South and the two other poorest regions? These have less government spending with respect to their respective regional GDP

The figures at hand are not gross taxes however; these have been netted with subsidies (our Compensation Fund) which makes computations even easier; indeed, national accounting equalities tend to assume perfect funding from taxes to pay for government expenditure. Poorer regions – in this case, the bums are the Southern provinces, Taza-Alhuceimas and Tadla-Azilal would share their output between consumption and government expenditure. This is precisely the case for Taza and Tadla, where Investment per Capita (and at a smaller extent, Net Exports) make up for less than 2% of GDP per Capita. These two regions, by the way, should have received a net 1.5Bn dirhams either as tax cuts, or direct government transfers. But they did not: the local population had to make do.

On the other hand, the Southern regions are a riddle when it comes to national accounting; its taxation is a record low, and the assumption behind national accounting does not stand. And that is so because the tax aggregate used for that matter was net of subsidies. Think of it as a reversed budget balance: G – T instead of T – G. One additional step would be to propose: T - (G_0 + G_s) where G_s is government subsidies expenditure. The balance is the net government transfer the region benefits from.

So what is the score? It is always difficult in view of the numerous shortcomings of proposed methods, but it is clear the remarkably high Southern GDP per capita (30,000 dirhams) which marks these regions as the third richest is solely due to large government transfers, in this case 7.2Bn dirhams in 2011 – .89% of GDP, 17.1% of subsidies dispatched to 3.5% of total population. The bums in this case, those who benefit from government transfers, are the Southern provinces.

Regional solidarity is an admirable principle, and should be encouraged at every level of government business. But it assumes transparency in these transfers, and some kind of economic logic to it. In this case, transparency is a vain word – let us not forget the assumptions behind all these computations are very formal, and that means reality might be a lot dimmer, i.e. actual transfers are higher. And the proposed newly redrawn regional boundaries will certainly not help.

The political ramifications of unequal and unjustified (from an economic point of view, anyway) government transfers from hard-working citizens to others will exacerbate resentment, and there is no doubt unscrupulous politicians will seize upon this if and when an electoral advantage would weigh in. Another way to look at it is instead to push for larger devolution; fiscal autonomy would then show how each region actually does in terms of economic performance, and a dedicated federal fund can then be set up to support those regions with structural difficulties, on the grounds of economic support, not back-room political strategies as it is now.

Advertisements

The Yield Conundrum

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on August 22, 2012

I have been going on and on about Morocco’s public debt for quite some time, and I must confess some degree of bad faith in all devoted blogposts. And then, as usual, twitter-conversations have guided my interest; Nadia Lmlili and Ahmed Benchemsi kindly suggested to devote a simple piece about the public debt, and esteemed fellow blogger, Anas Filali and I clashed over the impact of public debt on households, more specifically on the younger generation: he argued inflation and uncertainty need to be accounted for, I fired back it was a blatant sign of economic illiteracy to suggest so, since interest rates already bear that (coming from an assumed PhD candidate, and an MBA graduate, it is rather hilarious)

I have now the opportunity to address both issues; please bear with me on some details because it may get technical at points, nonetheless, I shall do my best to bring simplicity to bear.

The yield on public debt is in simple terms the interest paid on the bonds issued by the treasury. Governments everywhere do not fund themselves exclusively with taxes and other duties paid for by taxpayers; they also borrow money, with different maturities: sometimes, ministerial departments need money to pay for routine expenses, such as paywage; when long-term investments are carried out, adequate financing sources are sought. Is it a good or a bad thing? Economists do not always agree on the subject, but there is good evidence to suggest public debt is actually a net wealth; As far as Morocco is concerned, what I find backs the idea that government intervention (through taxation and borrowing) tends to affect favourably investment fluctuation (that is, the mere existence of government expenditure/taxation brings investment volatility to significantly lower levels). Let us therefore posit from now on public debt is, in principle, a good thing. What might then be argued is: “how much public debt is good a thing?”. For emerging economies, the figure of Public Debt/GDP ratio of 60% is considered to be acceptable; for a small, outward-oriented economy that is Morocco, debt, foreign or domestic, cannot be indefinite, and thus the proposed threshold is a good benchmark.

I would argue the public and the mainstream journalists do not bother with such indicators. I mean, from what I read in the newspapers, economic-oriented pieces are either literary comments of official documents, or usually inaccurate interpretations of these figures.the recent PLL deal with the IMF, as well as the eminent new bond issue for 1Bn dollars pushed public finances and debt front and centre. Unfortunately, journalists and bloggers alike were too quick to summon the painful memories of the 1980-era of structural adjustments and hardcore austerity. I wish my voice would carry more weight and say: stop the frenzy, we are not there yet, and the road is long enough to adjust with moderate austerity costs.

I said I was implicitly telling half-truths about the debt in Morocco because I focused only on absolute values, namely the debt stock and annual PSBR (Public Service Borrowing Requirements) and I was too quick to forget the useful teachings of E. Fama and the general equilibrium models. My primary mistake was not to look closely at interest rates paid on these borrowings and the piled debt. In my defence, theses figures are not easy to come by, and their span in time is limited to the 1990s, so I shall make due with what is available – namely the synthetic yield per maturity listed on Bank Al Maghrib website.

Let me describe what I want to do by first describing the Fama-Arrow-Debreu theory applied to interest rates: these are supposed to embed every kind of information pertaining to its class asset; the yield on public debt is supposed to signal, among others, the robustness of Morocco’s public finances, the expected return from the projects the money is spent on, the credibility of the treasury to pay back the debt when it matures, and expected inflation when such time arises. In my argument therefore, I assume the interest rate to be a perfect signal, in what is called the Weak form of Market Equilibrium; weak because it assumes no unforeseen events, and all available information is already taken into account. For those interested in reading about the theoretical argument, you can read this great paper by Fama.

We now proceed with the basic idea in this blogpost: the monthly variations in estimated yields of various maturities reflect agent anticipation of risk and (potential) default; high yields do not mean intrinsic returns increased; in fact, long term interest rates remain very stable, but the risk premium varies following agents risk aversion vis-a-vis sovereign debt. Risk premium is exactly what we are talking about, as this captures all elements mentioned above many agents take into account when they go on the market and buy treasury bonds. That risk premium is difficult to assess, as far as the simple method to determine the yield curve goes – it can be deduced from empirical data, the difficult resides in assigning the proper model specification to it.

Consider two maturities of 1 year and 5 years. On May 2001, their respective yields were 4.99% and 5.84%. You immediately notice the significant difference in yields, with the longer maturity paying off higher interest, precisely because the uncertainty attached to it is higher. After all, it is easier to forecast the state of government finances in the next two years than it is in the next 5 years. Also, inflation expectations are higher because prices will rise anyway.

We could also link both maturities as follows: (1+r_t) = \left(1 + \frac{r_t}{T}\right)^T

this method, when applied to the example at hand, results are remarkably close:

         |  1Y |  2Y |
---------------+------
  Actual |4.99%|5.22%|
---------------+------
Synthetic|5.03%|5.06%|
----------------------

The differences should not arise, since rational investors are basically indifferent between 1-year and 5-years bonds. But, in real life, these investors have certain preferences (the so-called market segmentation theory) and from this simple example, it is obvious in May 2001, investors preferred more liquid debt -that is, short term 1 year yield, hence the lower yield. Conversely, these investors require a higher yield for 5-years bonds because of their preference for immediate liquidity. This is a great example to describe risk aversion.

sudden dips in yields are due to the discrepancies in computations (spreads in risk premium) as many month were missing out.

The dearth in data on standard maturities (say from 1 to 15 years) is a blessing in disguise, because it allows for a practical application of the fancy fairy theory to hard stock reality. As usual, results are ambiguous; we observe indeed the same phenomena described above occurs from time to time, with significant spreads with respect to close levels. For instance, much of 1995 gainsays this because theoretical levels are much lower (with differences as high as 2 percentage points) than those observed for available months. The 10-year bond required a yield of 10% in September 1995 (remember those were the days of post-PAS, fiscal consolidation and low growth) even though theoretical yields computed on the basis of 1-year debt, closer to 8.3%. Why so? As mentioned before, Moroccan investors are quite risk averse, and prefer to invest in more liquid debt, in this case, debt with maturities shorter than 1 year.

Still and all, the constant fiscal conservatism observed over the past 20 years brought all yields down to historically low levels: my preferred benchmark, the 10-year bond yield, was cut down from almost 10.5% early 1995, to a little under 4.5% in 2012, that’s almost 2Bn dirhams saved on interest over the last 5 years. This is also true for other maturities (as shown on the graph); similar levels of high interest rates were observed during the 1980s, close to 8% for prime rates. This is why I do not share the alarmist reports on a new “PAS”: we are not nearly as bankrupt as we were back in the 1980s, inflation is not nearly half that of the past period, and domestic, let alone foreign debt is way below 1980s levels. That is why I confess my failure to report on that bit of news: compared to the 1980s and 1990s, present Morocco is a beacon of fiscal prudence and restraint, and have the yield curve to show for it.

We are not out of the woods yet, unfortunately. The profligate expenditure on food subsidies and the post-Feb 20 generous upgrades in civil service paywage, unmatched with new fiscal receipts, and exacerbated by the 2007-2008 tax cuts, resulted in increased borrowings. The result has been swift: post-2007 10-years yield increased 500bps (or .5 percentage point) the same can be said of shorter maturities, so this is no sunspot easily brushed off. This generalized push on treasury yields upward has been observed in recent treasury surveys, and does not seem to bother that much policy-makers.

In a sense, this was to be expected: the Banking sector is short of close to 50Bn dirhams in the past 18 months, and shrinking liquidities mean investors make choices and require moderately higher yields, including those of the public debt, not to mention the rapid increase in PSBR. In a perverted way, investors are willing to require a moderate increase in required yield for treasury debt to place their dwindling liquidities, rather than invest them back into a morose exchange. This is the conundrum: why would investors do that? I am afraid I cannot provide suitable answer on that one.

Back to the public debt, the upward yield push in all maturities means these investors expect inflation to increase over the immediate and longer term, and their expectations are partly vindicated by the latest HCP and IMF projections. Let us not forget government finances benefited from the 15-years historically low inflation (observed elsewhere in the world but it is particularly true for Morocco) to switch gradually its debt structure to rely almost exclusively on domestic finances. In this particular sense, there is still quite comfortable room of manoeuvre for the government to keep piling the debt on before it even reaches 2002 levels – that is, 10y yields at 6%. But then again, one needs to account for the diminishing risk premium between 2000 and 2007, and its widening ever since.

Are investors putting more risk premium on the benchmark maturity, or do they require less with respect to their liquidity preferences?

Let us instead posit the risk premium is only the remaining term from the following equality, for some yield r_t

r_t = R_p + r_0

where R_p is the risk premium and r_0 the long-term, risk-free historical return. Let us also assume this long-term guaranteed return is close to 1% (this is a fairly realistic assumption) then we observe it a good fit: indeed, while required risk premium has declined significantly over the past 20 years or so, the trend weakened and remained more or less flat since 2007, and remained around 4.5% for the benchmark maturity. I would suggest government talking points about the deficit carry little weight in prompting investors to bring bond yields back close to the 4%, because the measures proposed to match the objective of 5.3% deficit relative to GDP in 2013, and 3% by then end of 2016 are two impossible things to reconcile.

The last remaining question, but not the least, is: “are public finances in trouble to warrant strong austerity?” undoubtedly, no. But these are equally unequivocal signs credible measures need to be taken to reduce the deficit amid shrinking liquidities and troubling signs from the current account and the currency reserve. But we are still far away from the so-called ‘cardiopulmonary arrest’.

It is always difficult to convey cautionary tones, or even urge pre-emptive moderate austerity measures (from my own point of view, anyway) when both sides of the pond are screaming Armageddon. But the fact remains levels of debt stock are increasing at high speed, and though these do not translate immediately in increasing yields, it would be then a near impossible task not to pursue hardcore austerity to bring long-term yields back to 4.5%. It is even more disturbing to notice marginal spreads between short and long term maturities (less than 8bps between 1y and 10y yields recorded on July 2012). The matter of concern, in short, is not default risk; it is however that of a government increasingly unable to fund programs and investments because paid interest is higher each year are creeping on in.

The Economic Chronicles of the Kingdom, 1955-2011 Part.4

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on July 23, 2012

The Hauser Rule exists and it is verified in Morocco. Over a long period of time, almost 60 years in our case, the percentage of main tax receipts to GDP has remained constant, or at least did not rise above an upper bound, in our case, it is a little below 19.4% of GDP, except four years (and in good reason, as we shall see later on)

[…] En réponse à Horani, Akesbi précise que la pression fiscale ne représente que 22 à 25% au Maroc (il n’a pas décliné sa méthode de calcul) et que à ce titre, il est hors propos de demander des baisses d’impôt.

My Hauser boundary would at first glance contradict official figures from say, Bank Al Maghrib or the MINEFI. It would also contradict a statement from Prof. Akesbi, who puts his figure for fiscal pressure around 22-25% of GDP, which is probably true for the last three or four years, if all receipts except new borrowings are taken into account. It seems Bank Al Maghrib in the predictions laid in their 2010 annual report have made a similar assumption that public finances have been at their best in 2008 (a historical surplus in the Budget is indeed a plus) and the 24.2% should, if I am not mistaken, point to the total receipts (barring borrowings) relative to GDP. Unfortunately, both Prof. Akesbi and the BKAM team have missed the point of genuine fiscal pressure; the percentage is supposed to measure the treasury’s extraction of resources relative to wealth creation.

I do not buy into their argument for two reasons: first, the percentage itself provides little explanation as to its individual component, chief of which the contribution of taxes on Capital and Labour, and second, it gives disproportionate importance to various sources of treasury income with no immediate link to government and budget policy. The economic argument, the fiscal pressure should be computed so as to discuss the effects of distortionary taxes, and only then look at other lump-sum type of taxes, but certainly not pay too much attention to the miscellaneous receipts the treasury cashes in from the government’s portfolio, privatization or other minor sources of income.

Hauser said back in 1993:

The historical record is quite simple, if surprising. Not matter what the tax rates have been, in postwar America tax revenues have remained at about 19.5% of GDP. This is a lesson Congress should remember as it considers President Clinton’s proposed tax hikes. If history is any guide, higher taxes will not increase the government’s take as a percentage of the economy.

In relative size, the last decade has seen Morocco’s effective, “Hauser” fiscal pressure hover around 18.2%, the lowest effective rate since the days of fiscal conservatism and austerity of the Structural Adjustment Plan of 1983-1992.

Distortionary vs Lump-Sump: VAT is a distortionary tax, and so is Income tax. stamp duties and local government taxes, more likely to be lump sum taxes. The difference is two-fold: first, lump-sump taxes do not affect economic decision-making. Because economic agents are assumed to behave in a rational fashion, their optimized decision equate marginal effects, which means constants such as the lump-sum tax do not enter into account. This is perhaps why economists prefer them. On the other hand, these taxes are very unfair to the poorest and less-endowed economic agents. Proportional or quasi-proportional taxes are said to be ‘fairer’, but at the same time, they alter, or distort, economic calculations. In size as well as in importance, distortionary taxes are worth the study, while lump-sum taxes are a secondary element that need not be involved in the way described above.

the Hauser boundary evolves around 19.2% and 19.4% of GDP. This means fiscal policy has little impact on government’s fiscal pressure over the economy.

Some of the computations I present the reader with are built on a very strong assumption, almost a dishonest one: the only available data time series I could put together involved government expenditure, and not taxation. What you see really is government expenditure relative to GDP. My assumption postulates the primary balance is stationary around zero, i.e. government expenditure is almost fully funded by tax receipts. It is a strong assumption indeed, but some results from empirical data tend to vindicate yet again that assumption.

The years 1976-1979 stand out as a bit odd, because there was a primary deficit back then for obvious, historic reasons: huge transfers to the newly recovered Southern provinces, and a rapid expansion of Morocco’s military capacities have put a strain on its public finances. Besides, even though Morocco’s GDP grew at very high rates, it was not economic activity that pulled it off, and that might explain why it did not translate into additional tax receipts. Barring these 3-4 years, government expenditure and main tax receipts are not statistically significant, when one takes into account for instance the GDP deflator. For reference, I compare my expenditure-turned-tax receipts against the World Bank’s nomenclature GC.TAX.GSRV.CN and GC.TAX.YPKG.CN.

The assumption about the primary budget balance is one of long-term consequences: no country can afford a deficit in its primary balance, i.e. not fund its daily expenditure with taxation over a long period of time. This is particularly true for the past decade, where a primary surplus of 0.1% relative to GDP, on average, has been observed – this figure is merely the difference between distortionary taxes and government expenditure relative to GDP, and shows the long-term behaviour of public finances: primary taxation funds entirely government expenditure.

How should distortionary taxes have been levied? First off, we need to take a look at the long-term breakdown of production per input: if we restrict ourselves to capital and labour, total receipts from primary taxes should encompass the same proportions -captured by \alpha and \beta in Y = A K^{\alpha} H^{\beta} in order to neutralize the effect of exogenous technological process (captured by A) we assume \beta = 1 - \alpha. In this respect, growth gains and the respective contributions of inputs are distributed such: \ln(Y) = \alpha \ln(K) + \beta \ln(H)

In the realm of public finances, and with no loss of detail, labour taxes are levied on income, consumption-oriented goods and services (typically, VAT) while taxes on Capital are usually centred around corporate tax (a tax on profit or in accounting terms, operating margin). When one considers fiscal receipts from the last 20 years however, this does not seem to be the case: Capital is over-taxed, and Labour under-taxed.

the spendthrift late 1970s have ransacked the fiscal house, and impaired its stability for the next decade, and deflect it away from a 50-years mean of 15.6%

There are many ways to explain these discrepancies, both at the aggregate and input levels: first, fiscal policy in Morocco does not seem to take into account the repercussions of its implementation, meaning that the various tax breaks, deductions and even the new fiscal measures fail to anticipate the behaviour of agents subject to these fiscal regulations.

This is not a new phenomena, really: if indeed the Hauser boundary is verified, fiscal policy, translated into fiscal receipts, appears to exhibit higher levels of volatility -almost twice as much as GDP’s, though the trend observed since the mid-1990 points to a stabilization close to GDP fluctuations. The second point about these discrepancies is policy-making: the figures in this post fail to account for the differences in fiscal regulations, especially those pertaining to agricultural output, whose tax system has been frozen in effect since the mid-1980s. The same fiscal regulations miss out on the upper income bound due to the standard income tax, whose marginal rate actually falls when it comes to the top decile income earners.

In policy terms, income rates have been too low, or inadequate. The same can be said of consumption-based taxes, such as VAT. As for corporate taxes, though the effective tax rate is comparatively low, the receipts are not up to scratch, in terms of Laffer Curve, corporate taxes are not efficient, and need to be cut accordingly. To make up for the shortfall and to balance the fiscal ratio up, wealth tax and the agricultural tax need to be levied at some point.

The Economic Chronicles of the Kingdom, 1955-2011 Part.1

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on July 12, 2012

It’s been a long time I did not show up on twitter, and I am the better for it. I wouldn’t pretend I’d missed the impassioned twitto-debates I have been involved in, but in matters of economics -as much as in other areas- complex issues cannot be boiled down to 140 characters.

Summertime and with it comes time for self-assessment. A lull to reflect on the past year.and I would like to devote part of my time to a dear project of mine, one that I would like to devote my professional life to: a comprehensive chronicle of Moroccan economics, from the mid-1950s up to date. This I would like to do because the mainstream documentation on that subject and other contiguous ones is not, to my opinion, satisfactory. This is by no mean unfair criticism of their authors, often reputable scholars that however fell victim to trivia and the politically correct. But the fact remains, there are very few papers I can lay my hands on that resemble anything close to, say whatever Lucas, Sargent, Prescott, Kydland, Blanchard, Krugman (to name but a few über-heavyweights) produced in macroeconomics. I am sorry the Cinquantenaire report and the paper commissioned by the Abderrahim Bouabid Foundation did not rise up to the challenge. At least I will not disappoint as a pseudo-amateurish academic blogger, will I?

I suppose it is extraordinarily pretentious of me to even assume I can write up Morocco’s economic history. And surely it is. But my amibition is otherwise: I wouldn’t like to merely report on economic facts and statistics – for that and many other useful comments, please refer to the never boring Annuaire d’Afrique du Nord. As a matter of fact, it is because I had the pleasure of reading some of their earlier issues on electoral results from the 1963 general election, that I though I can take a look at data circa 1950s-1960s and confort it to mainstream macroeconomic models: what went wrong, what went right. You see, the Moroccan economy as a whole tends to exhibit steady properties. That says a lot about some policies and their effect on the economy, but still. What an orthodox Marxist sees as the inevitable dynamic of History (capital H, as in Hegel) I would suggest measurable economic forces have been at work.

What development means: a simple case study.

The problem at hand as far as the Moroccan economy goes is how it fares relative to the rest of the world, and more precisely, how it does relative to advanced economies. I had the opportunity to present the reader with some blogposts on the matter, but back then I lacked the adequate skills to put a convincing argument to my claim. I don’t implying I fully master these skills, but the analysis is definitely getting better.

Development in its crudest definition means simply the (accelerated) accumulation of physical goods, or capital. One may disagree with it (if anyone is interested, Joe Stiglitz, Amartya Sen and JP Fitoussi produced an interesting report on Gross National Happiness) but as long as no credible alternative analysis to the theory of capital accumulation, we should go with the existing. So the faster an economy cumulates capital (in real terms) the better it is, and the more developed it eventually turns out to be.

Keeping up with pre-1977 levels of growth could have made Morocco 39.87 Billion dirhams in real terms, and double our GDP per capita by the end of the 20th century

But, in order to do so, one needs a yardstick to measure that progress. I would like to consider reusing the comparison to the United States, to which I would like to add South Korea in the mix, and start off with a graph depicting real GDP per Capita in these three countries from 1955 to 2000. Morocco’s GDP is supplemented with a fitted estimation of its annual growth with pre-19777 levels (and variance)

The graph shows compelling evidence as to how Morocco does with respect to that convergence theory: Morocco had, up to 1977, levels of growth fit to carry us to significantly higher levels of income, aggregate and per capita. Indeed, between 1955 and 1977, average real GDP per capita growth was around 7.4%, and decreases to 6.4% from 1977 to 2000, a full percentage point lost in 23 years.

This is so-called Beta-convergence: because Morocco was a poor country in 1955 and 1977 by any measure, it is assumed it will grow at a faster pace, so as to catch up to more advanced economies; by comparison, the US economy (per capita) grew 6.4% between 1955 and 1977, and the South Korean economy, 7.1%. If anything, up to 1977, Morocco was doing pretty well. This begs the question: what did go wrong? Why didn’t Morocco maintain its average growth pre-1977, and does it really matter?

Perhaps it is not all about higher levels of growth. Professor Najib Akesbi, an eminent contributor in both reports mentioned earlier, did confirm his preference for the Beta-convergence theory: He believes -correctly- the economy needs an annual rate around 8% to achieve levels of income observed in big emerging economies; actually, when demographic growth is taken into account (around 1%) the level of growth to which Prof. Akesbi refers is the magic, pre-1977 growth per capita.

Let us test that hypothesis: let us assume that indeed, Morocco has been growing at annual level of 7% between 1955 and 2000 at around 3.8%. We then look at the gap between synthetic and empirical data. Results seem to contradict Professor Akesbi’s claim: for Morocco to achieve an annual 7% since 1955 results in a net real gain of 17.1Bn dirhams, and real GDP per capita would only increase 60%. So the Beta-converge does not explain Morocco’s failure to establish itself as a promising and dynamic emerging market; it is not enough to achieve high levels of growth. Sure, it definitely provides a higher GDP per capita, but not up to levels worth supporting the case for unconditional high growth. Something is definitely wrong with that crude theory of capital accumulation. And I would like to read some paper from mainstream economists here on the virtues of the other way to achieve higher levels of output per capita: the Sigma-convergence, or steady reduction in output volatility.

Strong growth 7% loses 40% of its effect to historical volatility. That number alone does not mean much.

In statistics, standard deviation provides a measure of how far apart individual observations are from the mean. If standard deviation is high, that mean… means nothing. In our particular case, a high level of output tends to lose some of the accumulated effects of high growth rate: doing 6% on a row for two years yields 6%. doing 0% and then 12% yields a real growth of 5.8%, since:

y_t = \sqrt[t]{\prod_{i=1}^{t}(1+y_i)}

The standard definition of a geometric mean suits growth levels better because it is just another application of compounding interest rates. Besides, it also captures precisely the cost of an unconditional Beta-convergence, in our particular case, the 7% with historical volatility (standard deviation) advocated earlier turns out to be 6.6%, and .4% is not a marginal quantity, it amounts to about 2.4Bn dirhams, in real terms. These are lost because the economy cannot sustain itself at 7% over a long period of time.

This, in my opinion, is Morocco’s conundrum: it needs higher levels of growth than those observed during the last decade (and that explains why I restrained my comparisons to 1955-2000) so as to boost its GDP per capita. On the other hand, it cannot achieve it without high levels of volatility that would ultimately take away substantial amounts of that high growth. The balanced growth path needs to be found elsewhere: in each aggregate contribution to growth, among others.

Quarterly Data for the Moroccan Economy: A Shortcut

Posted in Dismal Economics, Moroccan Politics & Economics, Morocco, Read & Heard by Zouhair ABH on May 30, 2012

This is some heavy stuff: the post proposes to use statistical results from computations on cycles to develop, in turns, the broad aggregates and expand the level of details; we move from 56 years to 224 quarters (1 years = 4 quarters). Obviously, this gainsaid much of the computations performed in other posts: the HP filter does usually a lot better with quarterly data, and these do not come in hand – not so much a gainsay as it is a complete reformulation of any past doodling. On the other hand, the closer any official study got to it was the Finance Ministry’s 2009 study on Cycles, considered too short a time sequence – 1980 to 2008. This will explain some discrepancies in their results – it also explains why the Hodrick-Prescott filter works better, although I suspect any non-parametric filter would do equally fine.

If you torture the data long enough, it will confess.

(Ronald Coase)

Annual GDP is usually computed as follows: y_t = \sum_{q=1}^{4} y_q (adjusted for seasonal fluctuations). On the other hand, we might substitute this expression with a more probabilistic setting, with y_t the cumulative density for quarterly GDP (with a normalized annual GDP to 1) this statistical device is needed to generate the moments needed for further computations: mean, standard deviation and perhaps skewness. Obviously, this denies the modelthe possibility of capturing outstanding fluctuations, but there is an arbitrage to be made: scarce data, persistent fluctuations and the easy in computations.

Scarce data: it is a pity HCP or any other public institution does not publish long time series pertaining to the Moroccan economy – reliable data goes as far as the early 1980s, and annual data is usually not available on domestic sources (I have yet to find some lonely pdf file on the Finance Ministry’s website with dates from the 1960s…)

Persistent fluctuations: to capture these means eliminating outliers; no doubt the process will underestimate negative shocks because of the constraints put on the model.

Handiness in computations: there are some models available, for which assumptions -sometimes strong ones- have to be made in order to make sense of it all. In that respect, the results are not supposed to be iron-cast; in uncharted territories, it is hardly the case, and Moroccan business cycles certainly are.

The idea is to find a distribution across quarters so as to minimize differences in relative GDP over the considered period (graph depicts annual output)

From the beginning of the year till its end, growth is somehow randomly distributed across quarters. Perhaps ‘random’ is not the right word: what matters here are the moments of interest; it matters little to ‘guess’ the appropriate form quarterly growth displays over the very long run, and these errors will cancel out over almost 230 quarters. the initial step is to adopt the overly simplistic assumption that growth is uniform across quarter, that is, growth adopts the cumulative density of a uniform distribution, from Q1 to Q4, while the assumption is indeed very strong, it has the convenience to present us with smoothness: uniformity means inter-quarter growth will be very close to the yearly trend growth. In that sense, those recorded disturbances around the trend will be considered as the historical volatility we need to compute for more advanced distributions.

How do we check the results make sense? There is a benchmark out there for us to see: Quarterly data for US GDP per Capita is available, as well as the annual data for the percentage of GDP the Moroccan economy represents with respect to that of the US. A successful estimation for quarterly growth in Morocco means the relative GDP to the US should be very similar to that in yearly setting. The idea is simply to compute both countries’ trends, and from then on the ration across the available 196 quarters (usable US Data runs from 1955:I to 2004:IV)

We proceed to generate a normal distribution at each year, with N \left(\bar{y_{t,t+1}},\sigma_{229}\right) where \sigma_229 is a measure of volatility derived for the time being from levels of annual growth, and centred around an average measure computed on annual growth values. We thus obtain the following results compared to annual data:

                            GDPY
-------------------------------------------------------------
      Percentiles      Smallest
10%     6.252841       6.162857       Obs                  57
25%      7.11526       6.189924       Sum of Wgt.          57
50%     8.232385                      Mean           7.942243
                        Largest       Std. Dev.      .9970512
90%     9.121388       9.263059       Variance       .9941111
95%     9.263059       9.275018       Skewness      -.4843868
                      Quarterly_GDP
-------------------------------------------------------------
      Percentiles      Smallest
10%     6.259518       6.089419       Obs                 228
25%     7.096377       6.104532       Sum of Wgt.         228
50%      8.24034                      Mean           7.941979
                        Largest       Std. Dev.      .9909281
90%     9.101249       9.296097       Variance       .9819385
95%     9.253781       9.301066       Skewness      -.4863844
99%     9.296097       9.343733       Kurtosis         1.9077

As one can see, the quarterly data does not distort the original series too much, the trade-off being at the cost of a marginally smaller volatility and average GDP.

Brand new Morocco’s Quarterly RBC, and HP-filtered.

The newly generated doesn’t look much like the earlier results I posted on. Simply put, the data was not fit for cycle-generation. Some short-term fluctuations were not taken into account precisely because the data was annual, and subsequently, some outstanding quarters (in both ways) were skipped or instead exacerbated because of their effects on annual growth.

Overall, the essential features of ‘yearly’ cycles can be found in the graph too: boom-and-bust in the second half of the 1960s, the boom of the 1970s, and then the quagmires in short cycles in the 1980s and 1990s.

Interestingly enough, volatility is much smaller compared to initial projections; in fact, 38% smaller, as we can observe on the descriptive statistics:

    Variable |Obs      Mean     Std. Dev.       Min        Max
-------------+--------------------------------------------------
HP_Quarter_1 |228    2.28e-19    .048117   -.1296305   .1497003
HP_AnnualY_1 | 57   -7.30e-19    .078588   -.1399502   .1696304

(the difference in mean is not very significant, as both are very close to zero)

How does this cycle perform with respect to that computed by the Finances Ministry in their 2009 paper?

Les huit cycles d’affaires enregistrés durant les décennies 80 et 90 ont été marqués par la comptabilisation de 13 années de sécheresse entrainant de fortes oscillations de la production agricole et des secteurs de l’activité économique qui lui sont associés à l’amont et à l’aval. […] La phase expansionniste que connait aujourd’hui l’économie marocaine se démarque clairement de l’expérience des décennies précédentes […] Ce contexte d’évolution démontre distinctement dans quelle mesure l’économie nationale a réussi à amorcer un changement positif de structures économiques et à développer une grande capacité d’adaptation et d’amortissement des chocs. Les gains de stabilité et de durabilité enregistrés au cours de ces dernières années tiennent dans une grande partie à l’amélioration de la conduite de la politique économique et de la qualité des dispositifs institutionnels.

The findings about the expansionary cycle beginning from late 2000 are somewhat mitigated when one considers a longer time series: true, the average cycle has been less volatile (25% less than the 229 quarters-long time series) but these fluctuations are under-estimated in the MINEFI study because of the historical volatility they embed: the 1980s have been very volatile indeed, and the great moderation that followed makes cycles in the 2000s very moderate, hence the optimistic view held in the ministry’s findings. On the other hand, it seems the last 40 quarters have exhibited historically low volatility, which, when combined with those in the period 1980-2000, can be under-stated.