“Everyone wants to live at the expense of the state. They forget that the state wants to live at the expense of everyone.” (Frédéric Bastiat)

With the advent of democratized technology, there is an increased correlation among


people, corporates and countries at large. Nowadays, it is very difficult to live in its own ecosystem, spared by the ripple effects of an event which might occur elsewhere. The mechanism of “creative destruction” is happening at a higher pace but with the only difference that destruction is happening much more rapidly than the creation phase. On one hand, predictive and repetitive jobs will most likely be replaced, with time, by artificial intelligence systems and on the other hand, the economy is gradually becoming a platform one.  The slogan is to become light and flexible. For example, any government in place should be triggering its educational policy towards non repetitive and complex activities in order to be relevant in the future.

A culture which devalues manual labour and over romanticises elevated high school graduation rates which results in access to university for everyone will face difficulties in the long run. Mauritius, as a matter of fact, is always said to be undergoing major changes. Governments as a whole have been relentless working in order to produce a country with a 65% debt ratio, twin deficits, high rates of youth unemployment, a growing income inequality, an increasing land-based economy with undermines human capital investment, favouring economic rents of all types and a production machine extremely elastic to prices. Economic responsibility would suggest significant state and political reforms along with a questioning of how public spending is being deployed. As a result of that picture, the governments (recent past and the current one) have decided to embark in the enactment of a metro line (or tram) which would supposedly solve the traffic jam.  It is true that increasing the capital stock will enhance the economic capacity of the country and at the same time have a multiplier effect. But as any “rule” in economics, it depends on the environment or sample it is applied on. Therefore, Keynesian economics works in variable ways. Infrastructure spending will have positive effects on the economy as a whole but the question to be asked is if the Mauritian economy can support that kind of capitalistic investment in its current state. Also, knowing the track record of governments for such ventures, can we realistically purport that this kind of project will not increase the fragility of the whole country?

Significant infrastructure investments are known to be a norm in many countries. For some, it’s a sign of might i.e the Japan of the 90’s, for others it’s an inherent part of their economic growth i.e China. And with it comes the economies of scale which it provides, the short term indirect positive externalities it produces and the pride of “bigger is better”. Mauritius is a small island, so the relative size of such projects is significant, and for this rail project we are talking of 20 billion MUR worth of investment, financed partially by the Indian government.  Note that 20 billion MUR is approximately 5 % of the GDP. Financial literature points out the propensity of big capital investments delivering poor outcomes as a form of “fragility”, a term suggested by Nassim Taleb.  A thing or system which is said to be easily harmed by randomness is defined as fragile. Large capital projects deliver often negative net present value due to various sources of uncertainty that impact them during their long gestation, implementation and operation. In mathematical terms, there is a non-linear exposure to uncertainties that eventually deliver poor returns, which outpaces the stated economies of scale. Trying to define what we meant by large projects is difficult but the main factors which coin a project as big are :

  • Physical proportions
  • Inputs that are required to build the project
  • Amount of financial resources to be mobilized
  • Quality of the service that the project would provide (functionality, speed etc)
  • Time to build the project
  • Complexity of the project in a technical standpoint
  • Impact in terms of the number of people which will benefit or be harmed by it. It also encompasses the environmental aspects.

These factors are intertwined and some of their internal interactions are simply impossible to foresee.  Consequently, Nassim Taleb proposes to construct a fragility-antifragility framework to capture the type of randomness and risk of any venture. The example of a porcelain cup can be taken, which is vulnerable to a wide range of uncertain events. Fragility is typically irreversible. Once broken, the fragile cannot be readily restored to its original function. A thing (material, system, process, or network) has a unique “fragility threshold”, which careful structural analysis and experimentation can reveal, at which it will break down under influence from an adverse event known for some as a stressor. A third component of fragility is that as the project or system gets bigger, the stressor needed to break it will decline disproportionately.  So any project with a very low fragility threshold and a low recovery hurdle should require a great deal of cushioning to protect it from breaking.

The Mauritian Project

The total cost of the project is estimated around 20 billion MUR and the Indian State is granting 9 billion rupees. The rest would be a debt financing structure. Due to a lack of data, it proves to be complicated to model a framework in which we can realistically come up with tested numerical conclusions on the fragility of this project. There is a very simplistic linear econometric model reported in a paper (Envisioning a new public transport system for Mauritius), but the author adopts a non-stochastic linear framework for modelling such project, this is unfortunately very limiting.  In order to encapsulate the fragility of the system there should have been a clear document depicting a realistic grid of cash inflows and cash outflows over 20 years with a non-linear modelling of passenger flow. This would have helped to calculate the net present values and the costs associated. Then, a gain to pain ratio could be determined. Normally, this ratio is used as a rule of thumb (heuristics) in order to determine the fragility threshold. Whether this ratio is low or high would have helped to determine the sensitivity of the project. These figures are very important because of the lack of repositioning if somehow “stuff” happens and “stuff” will happen. For example, let’s say during the construction phase there is a misconception or an exogenous event which affects greatly the project, it would be impossible to stop it as these ventures are binary in the sense unfinished infrastructure is useless. So, completing the project will demand “whatever it takes” logic to finish the venture. Knowing the balance sheet and the state of our national accounts, do we have the luxury to adopt such stance? All these factors along with local ones like track record of governments in such projects would not only increase fragility but will tie up future governments with significant constraints; economic, political and geopolitical. Albeit, constraints are manageable, it should first be acknowledged with proper risk management and government should be transparent about it. Taleb depicts this phenomenon as a concave curve which shows that consequences can be out of proportions.

Source : Taleb

This diagram shows the non-linear consequences if ever an event should happen (sliding to the right of the variable axis). We can see that losses can be very important especially in an environment like Mauritius, a country not considered as virtuous when it comes to fiscal discipline and overall personal responsibility.

An international comparison

The lack of data for the Mauritian project unfortunately prevent anyone for an in-depth analysis. However, we can compare the local project with international standards even if it bears some flaws like the specificity of the financing and the detailed contours of the project which are unknown. When it comes to the issue of cost overrun, Bent Flyvbjerg, a specialist of the matter, calculated on a sample of 100 rail projects, a gain to pain ratio of 1.44, which means that the net present benefits exceed the net present costs by 44%. But more than 50% of the sample had a cost overrun of over 44%. We know that the construction phase will be spanning over a general election. Undoubtedly, this matter will be fully exploited by the political parties on matters such as the costs, the negative externalities caused by it. This would create a non-favourable environment for proper project appraisal and end up in the classic argument “he said, she said”. Having a general idea about the cost overrun is not an easy task. Let’s assume that Sweden is our “gold standard”. This country is known to be fiscally very conservative, public funds are constantly under thorough oversight by strong independent bodies as well as any public tender process. The country is also ranked 4th in terms of Corruption Perception Index with overall score of 88(100 being the least corrupt state) meaning it has a high sense of responsibility when it comes to government spending.  There have been 65 rail projects during the period 1997-2009 and the results are as follows :

Source: Swedish Road and Rail Administration

A 21% cost escalation was incurred on average with a 50% standard deviation which shows that even in a very well managed country, with a long rail construction history, cost miscalculation can be huge with a significant in-built volatility. Bearing in mind, that Mauritius is far from this country in terms of fiscal discipline, independent controls for the use of public funds, we can realistically suppose that stated or unstated cost overrun will be significant. Any statement short of that is an illustration of indulging in mere fantasies.

Multiplier effect?

A rise in government spending is often said to have positive effects on the economy through the multiplier factor which is simply the idea that someone’s incremental cost/expense is someone else additional revenue. So this process which is looped several times will give a boost to the economy. In economics, this is often stated as a “rule” when it’s not really one. There are very few academic studies which analyses the real effects of government spending as it’s a chicken and egg story. It is tricky to settle which come first; an increase in spending generates an increase in economic activity or the inverse.  With this amount of infrastructure spending, there will be most likely a boost in economic activity but research showed that the efficacy of spending exists when these factors are present:

  • Multiplier effect works best when median revenue is low
  • Low Level of Corruption i.e cronyism, nepotism
  • Structure of the economy (Labour Laws, investment capacity, savings level)

From the data available, it can be said realistically; that the metro project will have a positive multiplier effect if the above 3 factors will play in a positive manner. Mauritius is also a land-based economy, this will create some spill over effect on real assets which in turn will play as collateral for credit enhancement for some other projects. Unfortunately, knowing the kind of development happening in Mauritius, it is most likely that investments which will be done won’t be pushing forward the potential economic growth even if the overall infrastructure capacity will be enhanced. Regardless of partisan opinion on this matter, these windfalls would be undermined by unintended factors or randomness. The fragility embedded or its building up will have serious negative impacts on the Mauritian at large.  The various examples around the world shows that whenever there is a big investment in a country with low to average income, the economy tends to inherit the fragility of the project, for example, accumulating debt which is technically deferred taxes. It is well known that countries with insufficiently large balance sheets would endure the greatest difficulties to absorb outcomes of big bets. Unfortunately, this whole project can end up into a pyrrhic victory.


It is very difficult to predict what will be the outcome of such a project as there is no real consensus on the matter. Having a rising capital stock is beneficial as theory suggests that in any production function there are two main variables at play; capital and labour. Capital flows go where the rate of return is the highest. So, is the rate of return of this project high enough? We are in country, where bi (or >2) partisanship on these matters is inexistent and tomfoolery is a norm. This adds to the fragility of the project. Furthermore, there are also some cognitive biases which should also be taken into consideration. It is often known that experts fail to detect forthcoming issues and thus are prone to errors. Thus, it can lead to underestimation of the risks associated. Although, there is a natural aspiration for innovative and efficient ways of transportation in order to facilitate connectivity, a transparent cycle of reflection with confronting ideas is recommended. Do we need to slice the projects into smaller ones in order to have a sense of control? Have we explored all other alternatives? What types of cushioning do we have against adverse effects? Is there a real option analysis to manage the uncertainties? Have we done the necessary scenario analysis when it comes to the profitability of the project? What are the insurance mechanisms if an adverse scenario happens? All these questions remain unfortunately unanswered.

The opinions expressed in this article are those of the author. This article is not intended to provide an advice or guidance with respect to the matters addressed.