Nova workboard

a blog from young economists at Nova SBE

Enriching Lisbon while it sleeps

With the local Portuguese elections of 2017, a heated debate on the Lisbon touristic municipal tax incidence arose. This tax is decomposed into an overnight tax (a 1€ extra charge per night introduced in 2016 with a maximum of 7€, yielding 13.5M€) and an air and maritime arrival tax (introduced in 2015 only in the airport and materialized on a 1€ extra charge per arrival, yielding 3.8M€ and borne by ANA, which refused its prolongation to 2016). Basically, the arrival tax applies to all citizens except Lisbon residents, children under two years old and passengers having their connection flight in Lisbon and the overnight tax to every citizen above thirteen years old.

The effectiveness of the lodging tax has been recognized even by the political parties who voted against it. However, improvements can still be achieved. One discussed proposal is to increase the tax to 2€/night to guarantee more affordable housing opportunities, though its opponents defend that Lisbon cannot rely on the touristic tax to solve its structural problems. This critic seems reasonable because when taxes increase by 1% (perceived by consumers as a price increase), guests´ behavior may change (lodging demand elasticity) – empirical data shows a fall on the occupation rate. Nevertheless, the more inelastic the lodging demand (insensitive to tax increases), the higher the success probability of such policy.

If the local government´s objective is to increase the revenue levied by this tax, then two less obvious alternatives can be explored. Firstly, following Ancona´s (Italy) example, Lisbon´s city council could set a fixed tax independent of the accommodation´s length. Given an average stay in the capital of 2,36 nights and assuming no substantial future changes, setting a tax superior to 2,36€/person would expectedly generate a higher recipe. A possible critic here is the higher the number of nights spent in Lisbon, the higher the benefit acquired so discriminating the overnight tax according to benefit levels delivered to the tourist seems reasonable (benefits principle).

Secondly, note Paris´ example where lodging taxes vary according to the accommodation´s number of “stars”. Since hotels with five, four and three stars lodge most tourists in Lisbon, by setting a tax above 1€/per night for these accommodations and a lower tax for the remaining (tourism progressive taxation), the town hall could expect to increase overnight total tax revenue. Additionally, bearing in mind equity considerations, this proposal is justified by the typically higher predisposition to pay a superior tax of a tourist that stays in a three to five-star hotel than the one who stays in a hostel (diminishing marginal utility of income). However, this alternative presupposes not a fall on overall consumption but rather induce high spending tourist to “reduce only their savings”.

Personally, such tax makes absolute sense as to improve the quality of the city (not decrease the number of visitors), since the increase in Lisbon tourism is exerting higher pressure on existing resources. Accordingly, the proposed alternatives would be more efficient to achieve the desired goal while generating a higher revenue.

Margarida Castro Rego | 23848


Italian citizens are stuck in the gasoline puzzle

Since early 70’ Italian citizens have been suffering from one of the highest gasoline cost of the entire world and, after the powerful earthquake that hit the central region of the country one month ago, a new rise in the excise taxes on oil is considered almost sure by many opinion makers.

Compared with Germany and France, the major European economies, Italian gasoline’s cost for consumers is 10.7% greater: 1.31€/l for both German and French people versus 1.45€/l for Italians (data from are referred to the 20th June-26th September 2016 period on a weekly basis). Such a high level in oil price is due to a fierce taxation system, mainly on the excise side. Nowadays, the Italian excise duty for leaded petrol product accounts for 0,7284€ per liter according to EY “2016 Global oil and gas tax guide”. Throughout Italian history, governments have used the excise on oil to collect money, easily and rapidly, in order to cope with exceptional circumstances or natural calamities or wars, too. For instance, taxpayers are still paying excise for the Ethiopia campaign of 1935.

On one hand, oil producers, who bear legally the excise, face no troubles in shifting the tax burden to gasoline users. On the other side, Italian consumers can hardly subtract themselves from this huge excise incidence. In fact, oil is Italy’s most important energy source (according to

A possible explanation on this inelastic demand may rely on the inefficient infrastructures and transportation systems. Looking at OECD 2015 data, Italian investments in inland infrastructure, a key determinant of performance in the transport sector which include both spending on new transport construction and the improvement of the existing network, is only 0,40% of the GDP. This data is consistently low comparing to the 1% on the total GDP spent in a country as France, and to the German expenditure over GDP, which is around 0,60% (source:

Thus, when a new excise is introduced, consumers who use gasoline to reach their work place by car, could barely change their behaviour and start using an alternative mean of transport, like train, bus or metro. Perhaps those citizens who reside in metropolitan city, namely Milan and Rome, will start going to work with bicycle if there would be cozy and safe cycle paths developed as in Netherlands. Again, the main concern for Italians is a lack of substitution effect which could break up this stuck situation.

Focusing the attention on a long-term perspective, Italy 2015 investment level in renewables has been 33% less than German level (source: and the trend is steepening in the last three years. Nonetheless Italian investments in renewable sector is, quite surprisingly, the second one regarding the European Union, slightly higher than Sweden. So if Italian governments will do the necessary investments in both infrastructures and renewables, the next generations might be able to quit from the gasoline puzzle and, who knows, maybe excise duties on petrol will be used in a better way: as incentive to choose the “green way” in spite of polluted one, no more to raise money quickly.


Andrea Zilio

Democratic Game

Revealed Preferences & Distortions on Public Choice

Voting is the most sensible expression of social will. It enables people to decide and reveal preferences about politicians’ acts, policies’ effects and different levels of public intervention. This post pretends to link some crucial concepts that we study in Microeconomics course, explain why sometimes democracy gives unexpected results and how important are the assumptions of revealed and well-behaved preferences.

Combining equity problems, behavioural distortions and market failures, the Public Choice analysis is one of the most challenging economic exercises which considers an interaction of three different levels:

  • Individuals who have preferences and want to maximize their welfare through public and private consumption (considering a budget constraint in which taxes are included as the indirect counterpart from public provision).
  • The Social Institutions that want to maximize Social Welfare and find an equilibrium, which can be different from the individual optimal solution.
  • Politicians, more benevolent or more selfish, who want to be infinitely re-elected by pleasing their voters (who can punish politicians if they deviate).

Although each politician has its own ideology and ethical belief, economists have tried to find a way to explain people’s desires and performances in the electoral game The rules are simple: citizens have different needs and they are willing to pay for welfare. However, payers and beneficiaries are seldom the same, moreover, free riders and externalities insist on stiffening our problem, which may lead to under-provision.

The most difficult analysis is to reveal preferences and to aggregate them in order to define the most important function, which tells us what people need and how they value it. In the real world, it is completely impossible to design a beautiful utility function or continuous indifference curves, linking bundles with the same value for citizens. Therefore, governments may ask voters to define their preferences (and the easiest way to do so is) by ranking bundles or options with the help of public surveys or referendums. Actually, politicians have the incentive to reveal citizen’s preferences in order to prepare persuasive speeches and move themselves to a good place in the Downs Avenue.

Nevertheless, even if we could solve all these problems and know exactly the marginal value of public provision for all individuals, democracy would still have distortions sorting out electoral choices and ‘right choices’. Therefore, it is relevant to present the most famous Electoral Paradoxes and their contradictory results:

Condorcet’s Paradox, the most popular of all, presents a situation in which a majority election over pairs of alternatives may have no clear winner. Suppose that A, B and C are possible choices and each voter orders his preferences as the following:

Captura de ecrã 2013-10-19, às 06.23.58

After beating the first winner (A), C becomes the final winner. However, would this result be the same if the fist election opposed A and C? No, because C would be the first winner and would lose against B (1+2 vs 3). In the Condorcet’s Election, the Voting Agenda determines the result (the winner is always the last voted choice).

Ostragorki’s Paradox involves the role of political parties as mediators of Public Choice. Supposing two parties (X and Y) and each individual voting for the party whose position is closest to his own:

Captura de ecrã 2013-10-19, às 06.24.06

Issue-by-issue, party X would win on every subject. However, if people choose parties to represent them, a majority of voters will elect party Y. This paradox presents one of the failures of the representativeness of parties.

The Additional Support Paradox considers strategic voting (of 21 voters) in a majority runoff election (the winner must have 11 votes, otherwise there is a second round between the two most voted).

Captura de ecrã 2013-10-19, às 06.24.14

In the first election, there is no absolute majority. Then, a runoff is held between choices B and C. It is easy to conclude that the winner is B (13 – 8).

Now, imagine that B receives the additional support from three ‘C-voters’. Still, we have no majority and the runoff will oppose choices A and B. Now, the final winner is A (11 – 10), despite of the additional support B has received.

When we think about Democracy, we will find many imperfections such as preference revelation problems, free-riding effects, electoral paradoxes, imperfect information problems, crowd psychological effects and general lack of trust on politicians. However, more than failures, these are challenges to solve with perseverance and always keeping in mind Winston Churchill’s words:

“Democracy is the worst form of government, except for all those other forms that have been tried from time to time.”

Dino Alves  # 607

Tips and Taxes

As described in detail by Andrew Schotter (79), tipping is a social convention with economic value. In this comment, I intend to give a brief explanation of the economic gains from this institution and then, in the line of my previous post, present its cost, namely its tax implication.

Whenever regulating the exchange of otherwise trivial services (such as haircuts, meals, etc) would involve negotiating and writing excessively long and costly contracts between client and performer, tipping, by allowing  variable compensation that rewards quality, comes on its own by avoiding these high transaction costs.

For example, when entering a restaurant, it is neither feasible nor cost-effective to design and enforce a contract specifying unambiguously all the conditions of the deal (how many (and how tasty) french fries will be served, the gentleness of the waiter, the quality of the glasses and cutlery, etc). So, except for some characteristics (namely sanitary and safety conditions) that are regulated, we are unsure about the food and the quality of the service we will receive -and because there is no contract, there is no way of litigation either. However, if tipping is instituted (i.e. the waiter expects it and I feel bound to tip if satisfied) there is less need for the extensive contract because both parties implicitly agree that a fixed part of the meal cost will be paid in any case but that the other, the tip, will vary according to the client’s level of satisfaction. In the case of dissatisfaction, the client simply does not tip or pays a smaller tip (and usually does not look for further recrimination). Conversely, the waiter expects to be rewarded for good service.

Furthermore, tipping is an efficient mechanism for the allocation of workers as it promotes self-selection. Again, in the restaurant industry, because of information asymmetries, the employers cannot really be certain of each applicant’s abilities at the moment of hiring. However, setting a payment system consisting of a fixed wage plus a variable component (the tips collected), will dissuade bad waiters from applying, as they know they won’t be able to collect tips enough to cover equal their opportunity wage. On the contrary, good waiters will be keen to accept those jobs, as they know they will receive lots of tips.

So, we conclude tipping is in fact advantageous for the satisfaction of multiple parties in the economy and that it contributes towards efficiency. However, tipping comes at a cost too!

Following my previous comment on the possible decrease of the VAT for the restaurant sector, we can also identify another advantage of restaurants over similar businesses: tips are not taxed! Clearly, if tipping allows for lower prices (as part of the service cost varies according to the client’s satisfaction, and is only paid after the issuance of receipts) then it is reducing the base for VAT, especially when the waiter service is a major part of the added-valued of restaurants. But this does not end here: restaurant owners also pay lower IRS, (tax on income), on behalf of the waiter since this amount is calculated taking into consideration the fixed part of the wage alone. Similarly, lower contributions to social security are paid.

Then, if tipping reduces the amounts of all three taxes paid by restaurants, on top of their known ease of evading taxes (they merely have to declare stocks as having expired), we clearly observe an unfair advantage over similar businesses, namely those who compete for the same workforce but where there is typically no tipping (for instance, shop sales staff: identically not well-qualified but not tipped for their services). This tells us that the real VAT rate restaurants pay is in fact lower and further justifies the previous argument that lowering the VAT for restaurants only is conceding an unfair rent.

Moreover, if we say that, with the present crisis, tipping rates fell (as consumers try to cut spending to minimums), the unemployment increase in the sector cannot be fully explained by the increase in VAT that occurred in 2011.  As tips fell, some waiters left their jobs or demanded higher fixed compensations; this may have led to the closure of many restaurants whose service quality deteriorated or who could not afford higher costs. So, the impact of lowering VAT as proposed by the government will probably be even smaller.

From another perspective, we can look at tipping as an adjustment factor that allowed restaurants not to dismiss employees. After the VAT increase, people knew they had to keep paying tips as long as they were interested in maintaining the previous service quality level (this may be plausible if we believe that the number of employees per restaurant did not decrease much, in this case unemployment is mainly explained by the full closure of many restaurants).

Matilde de Vasconcellos 666

Fires in Portugal

Fires are a scourge that every summer assault the world. We focus on fires in Portugal, which is the country where this problem is bigger in South Europe (the total burned area over the total area of the country is bigger than in France, Italy, Greece and Spain).

First of all, we should understand the development of fires through time. Data about the number of fires and the total burned area (in hectares) from 1980 until 2010 was collected. The moving average of 7 years was calculated, and it was possible to observe that the number of fires and their dimension have been increasing during the years although a decrease in the last few years can be noticed; nevertheless it is important to point out that most of the decrease observable in the moving average charts are due to the fact that in 2008 there were very few fires. What can explain this increase is now the most important question. Therefore, we decided to investigate the relationship between temperature and fires.

After we discovered that this is not a significant variable to explain fires, we tried to look at the political agenda, to see if would affect the number of fires. To do so we ran a simple regression that turned out to be insignificant, but nevertheless suggested a positive correlation between fires and elections, and for that reason we suggest further investigation in that direction.

In a nutshell, causes of fires are still not identified, which makes it harder to create an effective policy. Nevertheless, Portugal has been developing several interventions in the forest sector, such as the creation of an environmental fund, the Fundo Florestal Permanente (FFP).

The FFP was the first environmental fund in Portugal. It was created in 2004 due to an increased concern about sustainable forest management. It is managed by Instituto de Financiamento da Agricultura e Pescas (IFAP), part of Ministério da Agricultura, Desenvolvimento Rural e Pescas, and it is financed mainly by a part of the tax on petroleum products: 0,005€/l of petrol and 0,0025€/l of diesel, limited by 30 millions euros/year.

FFP finances activities that develop its goals through subsidies, credit lines, guarantees and insurances. Its main beneficiaries are municipalities and other public entities responsible for forestry administration as well as associations of forest owners. Since 2004, the volume of subsidies increased rapidly, but from 2010 onwards it has been decreasing. We could not find data about the projects financed. This lack of transparency should be a motive of concern since it is not possible to ensure that this money is being efficiently used.



Maria Martins – 540

Sofia Amaral – 538

External Imbalance

External Imbalance[1]

Presently much is being said about the Portuguese external economic imbalance.  Probably due to the fact that this country is under a bailout program headed by a troika of international institutions[2]. But what is that external economic imbalance? There are several aspects that explain it, though the central idea is an imbalanced flow of money between Portugal and the World.

Think Portugal as a person. Different persons engage in exchanges in a daily basis trading goods, services and factors of production between them and so does Portugal with the World (in aggregate terms). In the end, after all exchanges have been accounted for, if Portugal sees more money going out than coming in of its “pocket” then it is “imbalanced”.

To understand this let us draw a simple model. As said before the individual of interest is Portugal. This “individual” engages in trade with others, which can be captured by its imports and exports. The latter being the way of generating income to spend on the former (the desired goods and services that other “individuals” produce). Time will be reflected in the two period logic of this model: Portugal will engage in trade both in present (year 0) and in future (year 10). This 10 year span really stresses out the implicit contract that each individual does when balancing resources between the present and the future.  Its “pocket” will be defined as the possible combinations between its total present value and total future value of income: Portugal either spends all its possible income in the present or it spends all its possible income in the future or something in the middle. An interest rate is needed to compute these values: at year 10, the total possible income will be that year exports plus year 0 capitalized saved exports and, at year zero, the total possible income will be that year exports plus year 10 anticipated exports at a capitalization cost. This interest rate is just an average of the 10 year government bond interest rate of Germany, USA and Japan. This interest rate specification is somewhat subjective, but is believed to have important properties for this model: its term suits our 10 year span between periods, it is robust because takes in account three of the most representative economies and it is risk free.

Applying this model to the period 2000\2010 renders a consumption bundle of imports of €51.9\€62.8 (Graph 1 helps to visualize it). In 2000 value, that bundle means €112.4, while the present value of income (2000 exports plus 2010 exports at 2000 value) totaled €85.4 generating a trade “imbalance” of €26.9. This can be found in Table 1 and for decades 1989\1999 up to 2002\2012. It is striking the persistent record of “imbalances” between Portugal of a given year and the same “individual” 10 years later. The explanation for this lies away of our simple model: flows of external credit have been allowing this kind of behavior. It is, then, no surprise why Portugal ended up to be intervened by the troika… Portugal, like anyone of us, cannot persistently run out of its “pocket” without consequences.

  Table 1

Year of Period 1 Year of Period 2 Average Interest Rate Present Value of Exports Present Value of Imports “Imbalance”
2002 2012 2,98% 92,5 € 105,8 € -13,4 €
2001 2011 3,81% 89,7 € 109,6 € -19,9 €
2000 2010 3,87% 85,4 € 112,4 € -26,9 €  
1999 2009 4,49% 78,3 € 106,2 € -27,9 €
1998 2008 3,58% 82,8 € 109,0 € -26,3 €
1997 2007 4,35% 79,9 € 101,3 € -21,5 €
1996 2006 4,99% 74,1 € 94,0 € -19,9 €
1995 2005 4,89% 67,5 € 88,2 € -20,7 €
1994 2004 6,67% 63,5 € 83,2 € -19,6 €
1993 2003 4,88% 60,4 € 77,4 € -17,0 €
1992 2002 6,22% 58,3 € 77,4 € -19,1 €
1991 2001 6,71% 56,0 € 74,6 € -18,6 €
1990 2000 7,86% 54,9 € 71,7 € -16,7 €
1989 1999 6,99% 50,1 € 66,2 € -16,2 €

[1] All monetary values are at 2005 prices, in€, rounded to the first decimal. Exports and imports come from AMECO. Bonds’ interest rates come from Bloomberg.

[2]  The IMF, the EC and the ECB.

João Firmino

Efficiency and the competitive equilibrium

The first theorem of welfare economics (FTWE), states that prices of the competitive equilibrium will lead to a Pareto optimal allocation. Disregarding the possible flaws in the Pareto optimal allocation, (social inequality for instance), with the assumption of agents being price takers and the completeness of markets we can guarantee the Adam Smith “invisible hand”, or at least a simplified version of it, without taking into account the other achievements of markets, as innovation or its growth incentives.

The graphical and mathematical prove is possible, even in a general equilibrium context and therefore it would be best for the market to regulate for themselves. But then again, the Greenwald-Stiglitz theorem states that for an incomplete market, or in an imperfect information context, markets are not Pareto efficient. So the question that imposes is whether is it plausible to assume perfect information and completeness of markets, and if not if this will actually change the result of the competitive equilibrium, as stated by the FTWE.

Well it is hard to find a market that fulfills the assumptions, as perfect information is a rare thing, and there is no big number of suppliers in many markets, (perfect competition markets are mostly ideological), nevertheless by the violation of these assumptions it doesn’t mean that the reverse of the theorem holds, that is, that markets can’t be efficient – the assumptions of the first theorem are sufficient and not necessary.

On one hand we have Stiglitz defending the need of perfect information and completeness of markets so that markets can achieve a Pareto optimum outcome, on the other hand works from economists like Vernon Smith defending that even in a market with few producers and irrational consumers, markets can be efficient.

Probably markets without satisfying the assumptions of the first theorem of welfare economics won’t be Pareto efficient, but nothing proves that they won’t achieve the most efficient outcome, as Pareto efficiency is hardly a achievable efficiency. The question that stays to answer is which is a policy more efficient than markets. It is not because something is not efficient that it is not the most efficient tool existing.

The reason why most of the times markets do not operate on their own is that the efficiency optimum is not the social optimum – the need of equality in the politic agenda is something that from a society point of view is extremely important, and the loss of some efficiency in order to get some equality is necessary. We can see equity as a benefit, and the loss of efficiency a cost, in economy this optimum should be when the marginal benefit of equity gain is the same as the marginal cost of the loss of efficiency.

But then again we have an additional problem – where should the policy stop and markets run freely? Now this is a question of point of view, not withstanding the fact that there will be lost of efficiency with overprotective policies.

Maria Roque Martins – nº540

Making Consumers Better-off: From Liberalisation to a New Tariff Structure?

The on-going liberalisation of the Portuguese electricity market has entered in its last phase and from January 2013 the prices of electricity will cease to be fixed by the market regulator in all consumer segments. Such liberalisation is associated with an expected increase of competition in the electricity market that can leave the final consumer better-off. But will this really happen? Tariffs can tell.


Indeed, in light of the limited number of electricity operators under the new liberalized market, increased competition can provide more hope than a hint that the new electricity tariffs will leave consumers better-off. Ensuring that the consumer will benefit under a new market paradigm could involve convincing operators to adopt a new tariff structure.


A swift overview on the way the tariffs are structured by EDP to one of the consumer group segments ( provides a quick insight on a key characteristic of electricity production: electricity is relatively cheap to produce – and sell – up to the amount of installed capacity. Above this quantity, the costs of generating an extra kilowatt are enormous. From the operator point of view, differentiating prices in peak and low electricity demand periods is thus an effective way to control demand and prevent the expensive costs of installing additional capacity. Yet, such tariff structure – which is widely used in the liberalized marked – could be friendlier to the consumer.


Indeed, charging a higher electricity price during peak demand than during low demand implies that the amount of electricity a consumer can afford during peak demand will be lower than the amount of electricity the same consumer can afford during low demand. For those consumers that can vary consumption across demand periods, this tariff structure leaves them equally-off as they will be able to keep their total electricity demand constant by opting to consume more during the low demand period and less during the peak. Yet, for those consumers who cannot avoid consumption during peak demand, they will be worse-off by seeing their electricity bill go up. Could a different tariff structure enable the operator to address the capacity issue while allowing consumers who cannot vary consumption to be equally well-off?


A possible solution could be a new tariff structure; one that would rely on a baseline quantity contracted by each consumer at a base price and representing his/her normal usage. In periods of peak demand, any consumption above the baseline quantity would be charged at a higher price, whereas reduced consumption would translate into a rebate on their final bill. In periods of low demand, additional consumption would be charged at the base price and rebates would not take place. This possible structure would leave those consumers that cannot vary consumption equally-off while leaving consumers that can vary consumption better-off. All of this, whilst enabling the operator to avoid the expensive costs of creating additional capacity. Would this be the best tariff structure for the operator? Possibly not, but at least it could increase his competitive position while leaving the consumers equally or better-off.

João Nolasco