On the 15th of January, the Swiss National Bank, decided to remove the cap on the Swiss Franc-Euro exchange rate that was fixed at 1.2CHF/ since 2011. I m not going to look at the financial reasons that have led the SNB to take this decision, but I m going to investigate how the new exchange rate will impact the labour market in Canton Ticino. I focus on this Swiss Canton, at the border with Italy, comparing the results with other cantons that have similar economic features. I ve decided to look at this area because it allows me to carry out a natural experiment. I will study the effect of the sudden exchange rate differential in a labour market characterized by three identical groups of workers: Italian workers living in Italy, commuting on a daily basis in Switzerland; Italian workers living and working in Lombardy; Swiss workers living and working in Ticino. The main difference between these groups lies in their purchasing power. I m going to analyze the effect that this sudden exchange rate shock had on output and employment levels in Ticino using the 3-equation model IS-PC-MR (Carlin et al.,2014 1 ), (without considering the Central bank reaction to the shock). Ticino will actually react better to the loss in competition than other Swiss cantons because of the more flexible nominal wages; Italian Daily Cross Workers Commuters (DCWCS) will keep high their real wages even after having their nominal wages cut by firms. Because there will be workers willing to work for lower nominal wages, the Italians, Ticinian firms will have lower costs, and in this way they will not shut down. It s clear that Italian workers will be preferred to Swiss ones, at least in the manufacturing sector, which will be the most harmed by the reduced competitiveness. The reduction in exports won t be as dramatic and as long as in the other regions where workers can t cut their nominal wages without cutting their real wages. Even if there will be a reduction in the numbers of Swiss people employed, because of their reduced employability, many Ticinian producers will be able to survive thanks to cutting nominal wages of the Italian DCWCSs, and in this way they will be able to save many other Ticinian jobs. Integration of the Swiss Labour Market Through the agreement on the free movement of people between Switzerland and the EU, signed in 1999 and effective since 2002, citizens coming from both areas, have obtained the right to choose freely their place of work within the territories of the contracting parties. The free mobility treaty has removed quotas in the labour markets and has removed the priority that native workers had in their own nation. It meant that the labour market of the boarding regions, Lombardy in Italy and Canton Ticino in Switzerland, became a unique and integrated labour market, as suggested by Maggi and Gonzalez ( 2002). 2 Figure 1: Ticino and Italian provinces of Como and Varese 1 Carlin, Wendy, and David W. Soskice. Macroeconomics: Institutions, Instability, and the Financial System. Oxford University Press, 2014. 2 Gonzalez, Oscar, and Rico Maggi. "Segmentation by skills and wage discrimination in a trans-border labor market." ERSA conference papers. No. ersa02p445. European Regional Science Association, 2002. 1
The Italian provinces of Como and Varese, together with Canton Ticino can be though as twins separated at birth. The latter, is now under the political and economic control of Switzerland, while the are the twins brought up by, Italy. Later in the paper, I will look at economic trend and it will allow us to judge which one between the Italian twin and the Swiss one is the lucky one. Across the border, people speak the same language, share the same traditions and have a similar culture. The only true difference between the twins lays in the legislation. Therefore, workers commuting and working on the other side of the border, don t have to face any kind of cultural and linguistic barrier that may negatively effect their employability and wage (According to Alberton and Gonzalez, 2004 3 wages should converge to a common level in both side of the border). Any difference we find in economic outcomes is not due to any personal disadvantage the worker faces in the foreign labour market, rather it lays in the wider economic trends of Switzerland and Italy and, consequently, of Canton Ticino and Lombardy. This is, in all effects, a natural experiment. Economic trends As the following data will clearly show, Lombardy and Ticino are characterized by similar Economy. At the same time, even if we have similar economic features, we see that the Bilateral Agreement results in migration singularly from South to North, i.e. from Lombardy to Ticino. For example, looking at the growth trends of both areas since the signing of the Bilateral Agreement on the free movement of workers in 2002, we can see a common path. Positive growth has been a constant fact for both areas; this positive rate of growth for Canton Ticino is consistent with the Swiss Confederation s rate of growth, while Lombardy seems to be an exception in the Italian economic situation. Figure 2 4 : Rate of growth for the 2 areas Another relevant issue that needs to be analyzed is the level of unemployment. We can see how Lombardy before 2009 (light green line), year in which the Italian economy as a whole suffered the most from the global crisis, experienced considerably low levels of unemployment, which were in line with the Swiss ones (blue line) and were almost 50% below the level we can see in Ticino (yellow line). From 2009 there has been a rise in the level of unemployment for Lombardy and in 2011 it overtook the level of Ticino. It s important to see how the level of unemployment in Ticino and in Lombardy, sharply rise in 2011, differently from what happens in the rest of the Swiss Confederation, where the level of unemployment is constant on the pre-crisis level. 3 Alberton, Siegfied, and Oscar Gonzalez. Monitoring a trans-border labour market in view of liberalization. Università della Svizzera italiana, 2004. 4 Swiss Data taken from USTAT, Lombardy Data taken from Annuario statistico regionale 2
Figure 3 5 : Unemployment levels in Lombardy (green), Switzerland (blue) and Ticino(yellow) Another important fact that we need to stress is the almost equal division of the labour force by sectors of employment. In the provinces of Como and Varese, there are almost 250,000 workers in the secondary sector and more than 380,000 in the tertiary sector. Quite similar percentages, with clearly different absolute number due to different size of population, are applied to Canton Ticino, as the following graphs show. Figure 4 6 : Occupation by sector in Como&Varese and Ticino These common economic trends in the two areas, and the complete openness of the two markets, may lead us to think that the market clears; people would move across the two sides of the border till they are indifferent on where to work. In reality, as previously stated, the migration is just from one side; from Lombardy to Ticino. It is the phenomenon of the DCWCSs, also called the frontalieri. Frontalieri - DCWCS, for salary reasons Their status as daily DCWCSs is recognized and regulated by special laws between Italy and Switzerland. In order to be subjected to this particular legislation, that regulate aspects like taxations, pension schemes and unemployment benefits, you need to live no farther than 20 kms from the Swiss border. In fact, as I ve previously written, there are only Italian frontalieri. Since the Bilateral agreement became effective in 2002, the number of the frontalieri has doubled, from 30000 units to more than 60000 at the end of 2014 and they now account for more than ¼ of the whole Ticinian Labour force, as shown by the following graphs. 5 Data on employment for Ticino and Lombardy come from USTAT 6 Data for Varese comes from Annuario statistico regionale, Data for Ticino come from USTAT 3
Figure 5 7 : Number of Frontalieri and Ticinian Labour Market The macroeconomic data I ve previously reported are not enough to explain this phenomenon. There must be something else that encourages every day more than 60,000 people to drive for so many kilometres and make them the most envied workers in the area. It is the Swiss Franc and the different purchasing power you have by spending a Swiss income in the relatively cheaper side of the border, the Italian side. Envied by Italian workers working in Italy By looking at nominal income, we see that there s an impressive difference between the average nominal income earned by workers in Varese or Como and the average nominal income workers would get in Ticino. If we consider, as an average nominal monthly income for an Italian worker in the manufacturing sector in Varese, the estimate of 2.150, equivalent to 2800CHF (assuming an exchange rate 1.3CHF/ ), the following graphs, helps us understanding how the frontalieri are advantaged compared to Italians. Figure 6 8 : Wages by qualification levels and by working permit Looking at the previous graph, we can see how the frontalieri have an immense advantage over their colleagues working in Italy. Even the least qualified DCWCSs, manage to have a considerably higher salary than Italian workers. (Partially) Envied by Swiss workers too If nominal wages for DCWCS are actually higher than the nominal wages they would get in Italy, we need to look at real wages to compare Ticinian workers with the frontalieri. What really matters is the purchasing power workers have. 7 Data on the number of frontalieri and on the division of labour by nationality come from USTAT 2013 8 Data on average nominal income by qualification and work permit come from USTAT 2011 4
CHF 9 Swiss 100 76.92 Lombardy 75 57.69 The table shows that in order to buy a basket of goods in Ticino you need 100CHF. The same basket of goods in Italy costs 75CHF. It means that life in Italy is ¼ cheaper than in Switzerland. At the same time, data (figure 5) show that the nominal income of Italian commuters is sensibly lower than nominal income of natives, and this actually compensates for the purchasing power gap. This higher nominal income is a double edged situation for Swiss workers, since they manage to keep good standards of living in the more expensive Ticino, but at the same time it could make them less employable, since their income results in higher costs for business. It contributes to create a situation of wage dumping, where the cheaper labour supply from immigrants causes native s wages to fall. Figure 7 10 : Ticinian labour market by sector and by nationality of workers For these reasons, we have seen how the Italian DCWCS have been taking a bigger and bigger share of the secondary sector, where their general lower level of education doesn t harm their employability chances. It has also had political implications in Ticino, where nationalist parties, worried about the possibility that DCWCSs pressure on both unemployment and wages, are getting more and more consensus. In February 2014, the result of a consultative referendum in Switzerland, about the possibility of re-imposing quotas on the number of foreign workers in the Swiss labour market, had the highest percentage of yes vote (68.2%11 of the voting population) in Canton Ticino, although Ticino is not the only Swiss Canton that is experiencing the phenomenon of cross border commuting.(in both French and German speaking parts of the Swiss Confederation, there s even a higher number of daily DCWCS) Exchange rate shock Since 2011 the Swiss National Bank decided to peg the value of its currency to the Euro at the fixed exchange rate of 1.2CHF/1. It created stability and certainty both in financial and labour markets. On the 15th January 2014 the SNB suddenly decided to remove the peg allowing the exchange rate to fluctuate. Immediately, the Swiss Franc appreciated against the Euro, peaking at 0.85CHF/1, before stabilizing at 1 CHF/1. People, living at the border between Italy and Ticino, rushed to change their money, buying the relatively depreciated Euros and spend them in Italy. This exchange rate shock had an immediate effect on the competitiveness of Switzerland: p * e Q = p 9 According to an average exchange rate for 2011 equal to 1.3CHF/1 10 Data on the division of labour by sector come by USTAT 2011 11 Swiss referendum on 9 th February 2014: http://www.swissinfo.ch/eng/ticino-says--basta---to-cross-border-workers/37943558 5
where Q is the real exchange rate, reflecting the competitiveness of a country, e is the nominal exchange rate (CHF/1 ), p* is the Italian price level and p is the Swiss price level. The Swiss Franc has appreciated against the Euro, therefore e, reducing Q, and the Swiss competitiveness too. For these reasons the Swiss economy is now out of the Trade Balance curve; it s importing more than exporting. This process can be represented as a shift along the Aggregate Demand (curve) on the 3 equation model, AD-BT-ERU; AD represents aggregate demand for home s good (in this case Swiss goods), BT curve represent the balance of trade ERU curves represent the equilibrium on Figure 8: Shift along AD, from A to B, competitiveness goes down This movement along the AD curve, due to e, has also effects on the labour market; it causes an upward shift of the PS curve, since real wages actually go up, due to the increased purchasing power workers now have. At the same time, nominal wages are at B, on the WS curve. (N.B: the distance between BC is equal to the distance between B C ) Figure 9: PS curve shifts up 6
Both the difference between real wage at C, on PS, and the nominal wage at B and the lower level of employment, N1, generate deflation. This can also be told by the fact that B is on the left of the ERU curve that pins a condition in which inflation is constant. It creates a downward trend for Swiss deflation, that is already low (-0.3%). Inflation moves from point A to point B. Because output is still out of the equilibrium point Ye and expectations are assumed to be adaptive, inflation will go down till the economy is back to an equilibrium. Figure 10: Effect on inflation due to the differential between WS and PS Over multiple periods of time, deflation has an effect on both nominal wages and prices. As we have p * e previously stated, Q =. Then, since p, Q starts rising, and the economy slowly starts moving from p B back to A (figure 11). It means that we slowly re-acquire competitiveness through lowering prices and adjusting nominal wages. Moreover, the increase in Q pushes PS downwards, till it will lay on PS. As previously stated, inflation will go down further, since the economy is still below the equilibrium output. The closer the economy gets to the equilibrium point, the smaller the decrease in inflation will be. 7
Figure 11: Increase in Q due to adjustments in nominal prices This is the theoretical path the economy should follow, but we have to consider that the extremely prolonged time of both high unemployment and low nominal wages (being on the left of the equilibrium point), may create a Hysteresis effect in the market. Workers are not happy of seeing their nominal wages cut and therefore they are less inclined to work. At the same time, firms close permanently since their expectation for the future are extremely negative because of the appreciated currency that makes them less competitive; the process of regaining competitiveness, through nominal wage and price adjustment, takes too much time. It means that many firms close permanently, fire employees and consequently reduce output. Both the cut in nominal wages and the loss of jobs due to many firms shutting down, have a negative effect in the labour market; it creates a high number of long term unemployed individuals that are not willing to supply their effort for lower nominal wages. Being unemployed for a long time leads to a progressive loss of skills and to an erosion of psychological attachment to working life. It makes unemployed people not to push enough pressure in the labour market. As a result, we have a higher WS curve, because of the lower pressure of unemployed on employed people. A shift of the WS curve, creates a new point of equilibrium in the supply side of the economy, making the ERU curve shift leftwards too. At the same time, the fact that many firms will actually shut down, leaving people unemployed has a negative impact on the AD. The reduced capacity of exporting of Swiss firms, results in a negative impact on the BT curve too. Both of them, in the end, will shift leftwards. 8
Figure 12: Hysteresis in the Swiss market. For Ticino, the situation is different. The Ticinian job market will react differently since it is much more flexible because of the DCWCS. As data previously showed, they are much more willing to work for lower nominal wages, because their real wages manage to be high thanks to the lower cost of living. Differently from their Swiss colleagues, they care much more for their real wage than for the nominal wage, and therefore are not discouraged by the cuts in nominal wages. Moreover an appreciated Swiss Franc, allows the frontalieri, to have a higher purchasing power, even higher than before. CHF 12 13 Swiss 100 76.92 100 Lombardy 75 57.69 75 If before, with the exchange rate at 1.3CHF/1 Italian commuting in Switzerland needed 57.69 to buy a basket of good in Italy, now, with the exchange rate 1CHF/1, they have 75, that are more than the 57.69 they need for the same basket. Italians are even more willing to supply their work, putting pressure on Swiss workers. As a result, in Ticino the WS curve won t shift at all because the DCWCs will keep a high pressure on local workers. It means that also the ERU curve will be were it used to be. 12 I consider an average Exchange rate for 2011 equal to 1.3CHF/1 13 I consider an Exchange rate for 2015 equal to 1CHF/1 9
Ticinian firms will substitute from Native Ticinian workers to DCWCs since they accept receiving a lower nominal wage. It allows a rapid adjustment of prices and nominal wages and it results in a faster return to the equilibrium point, avoiding the Hysteresis effect (Ticinian economy follows exactly the theoretical pact of Figure 11). The flexibility given by the frontalieri has an effect also on the demand side of the economy because it reduces labour costs for firms that will manage to avoid the big loss in competitiveness that other Swiss firms may face. Companies, instead of firing and closing, will assume mainly DCWCS, reducing their costs. Reduced costs mean more funds available for investments, therefore the AD curve instead of shifting leftwards due the lower competitiveness, will remain where it currently is 14. (Figure 12; blue colour for other cantons where there s no possibility of hiring DCWCs, black colour for the Swiss situation of equilibrium before the shock and for Ticino after and before the shock). Figure 12: Ticino reaction compared to similar cantons reaction. As we can see from figure 12, the Ticinian market ends up with the same level of employment and output. It reacts better than the less flexible labour markets we find in other Cantons, where both output and employment are at lower levels (Point A ). The flexibility given by the presence of the frontalieri often seen as having a negative impact on the whole labour market, is actually the reason why Ticino would react better. It s true that DCWCSs will substitute natives or, as the nationalist Ticinian propaganda prefers to say, steal natives jobs. However thanks to them the effect of the shock on the whole economy is attenuated. It s possible that avoiding closing other firms, have saved more Swiss jobs than the ones that actually are stolen by Italian workers. 14 Clearly we can t generalize for the whole Swiss market, but we can compare Ticino to other cantons that have common economic features. 10
Conclusions With this paper I ve tried to stress how the presence of these Italian daily DCWCS is actually making the Ticinian labour market more flexible and more able to react to this exchange rate shock. Their presence actually allows to limit the loss in competition that Swiss companies would have from a relatively stronger Swiss Franc against the Euro. The fact that they are willing to work for lower nominal wages, since their real wages are unchanged because of the higher purchasing power they have in Italy, allows firms to cut costs and survive, avoiding the Hysteresis effect. Nationalist propaganda will probably look at the fact that the much more employable Italians will steal Swiss jobs in this period of lower output due to lower exports; but I think that through both my application of this model and the use of data, in the future we will be able to see whether the presence of the frontalieri has saved more Swiss jobs than it has stolen. This will actually allow for a more sane discussion on the effect of migration in Ticino based more on actual economic data and less on slogans. 11