Exchange rate misalignment and economic growth

Exchange rate misalignment and economic growth Victor Ribes Segura [email protected] Supervised by: Juan Carlos Cuestas Olivares EC-1049 Bachelor’s Thesis Academic year 2018-2019, degree in Economics July 2019 Abstract This paper study the relation of RER misalignments for three southern countries and for three northern countries, obtaining the misalignments from a VEC model. Concluding that misalignments are bigger in the northern countries with one exception. The next step done is through a basic equation of economic growth do a similar VEC model, later with impulse response analyses the effect of misalignments in economy activity for each country. The result are positive effects in southern countries with the exception of Spain and negative for northern with the exception of Denmark. Finally mixing the two results the results are not clear because are contradictory results and is not possible say if undervalued exchange rate is good o bad. Anyways, concluding, the misalignments affect the economy activity and it is growth for each country so exchange rate becomes an important instrument to consider for the countries. 2 Contents 1. Introduction ............................................................................................................... 3 2. Literature review ....................................................................................................... 6 3. RER Misalignments .................................................................................................. 7 4. Country Growth ....................................................................................................... 15 5. Conclusions ............................................................................................................ 22 6. Appendix ................................................................................................................. 25 7. Data Appendix ........................................................................................................ 29 8. Bibliography ............................................................................................................ 30 Figures and Tables Figure 1. GDP per capita .............................................................................................. 4 Figure 2. HDI ................................................................................................................ 5 Figure 3. RER ............................................................................................................... 5 Figure 4. Spain eq RER .............................................................................................. 11 Figure 5. Italy eq RER ................................................................................................. 12 Figure 6. Greece eq RER............................................................................................ 12 Figure 7. Denmark eq RER ......................................................................................... 13 Figure 8. Sweden eq RER .......................................................................................... 13 Figure 9. Finland eq RER............................................................................................ 14 Figure 10. Misalignmets .............................................................................................. 15 Figure 11. Spain impulse response ............................................................................. 18 Figure 12. Italy impulse response ............................................................................... 19 Figure 13. Greece impulse response .......................................................................... 19 Figure 14. Denmark impulse response........................................................................ 20 Figure 15. Sweden impulse response ......................................................................... 21 Figure 16. Finland impulse response .......................................................................... 21 Table 1. VEC of RER.. ................................................................................................ 10 Table 2. VEC of growth.. ............................................................................................. 16 Table 3. South RER cointegration ............................................................................... 26 Table 4. North RER cointegration ............................................................................... 27 Table 5. South growth cointegration ............................................................................ 28 Table 6. North growth cointegration ............................................................................ 29 3 1. Introduction For centuries, the currencies of the world were backed by gold. This means, that a currency bill issued by a world government represented a real amount of gold that government kept in a vault. In the 30s, the US established the value of the dollar at a single and unalterable level: an ounce of gold was worth $ 35, the gold standard. After World War II, the countries signed the Bretton Woods pact, whereby the IMF was created. Through fixed exchange rates everyone knew how much gold was worth a US dollar, the value of any other currency against the dollar could be based on its value in gold. A currency whose value was twice the value in gold of a dollar, was worth, therefore, two dollars. Unfortunately, the real world of economics overcame this system. The US dollar suffered inflation (its value relative to the goods it could buy decreased), while other currencies revalued and became more stable. In the end, the USA they could no longer pretend that the dollar was worth as much as it had been worth, so its value was officially reduced so that an ounce of gold would then have a value of $ 70. Finally, in 1971, the gold standard was over and countries applied a flexible exchange rate. This meant that the dollar no longer represented a real quantity of precious material, and change the model to one where supply and demand adjust the price, and in some cases with central banks keeping the price between some values. After 1971 the flexible exchange rate became the most used type of exchange rate. Today, the US dollar continues to dominate many financial markets. In fact, interest rates are often expressed in US dollars. Currently, the US dollar and the euro account for approximately 50 percent of all the world's foreign exchange operations. Including British pounds, Canadian dollars, Australian dollars and Japanese yen, we have more than 80 percent of all currency changes. On the other hand, the last years the importance of the exchange rate policies was fundamental for many countries, and has become a huge debate. Countries like China are the example of this policies, which are related with current account surpluses undervaluation his currencies for gain competitiveness. As other emerging countries that adapt the exchange rate policies to reach the developed countries. The BalassaSamuelson effect relate the exchange rate and economic activity, where variation in exchange rates may have an important effect to economy. 4 In this paper, I study the relationship between real exchange rate misalignments and economic activity for three southern European countries (Spain, Italy and Greece) and three northern countries (Denmark, Sweden and Finland). The reason for choose these countries is compare two realities, during the financial crisis the difference between the north and the south has increased. In the north the standard of live is on the top comparing all the EU countries, while the south is on the bottom (with the exception of Spain that is in the middle), capital flight from the south to the north and the effect in unemployment was so much bigger in the south. Looking at the GDP per capita the difference is enormous, we see how Denmark leads the table of the northern countries with the highest values followed by Sweden and Finland. While in the south, Italy leads followed by Spain and Greece with the lowest values. Observing well the data we see how the distance is abysmal between these countries, where Finland has twice the GDP as Italy, where also Sweden and Denmark get high differences with Italy. Figure 1. GDP per capita The Human Development Index is other clear example between the differences in the north and the south. This index integrates life expectancy, education and per capita income to create a ranking of the countries. In the Figure 1, clearly, we can see the difference while the norther countries stay in the top with Sweden in the head followed by Denmark and Finland, the southern countries stay in the middle of a ranking with 58 countries. As Spain as in the head followed by Italy and Greece. 5 Figure 2. HDI First, using the real exchange rates, in the Figure 1 are represented for the countries selected since 1995 until 2018 in quarters using the 2010 as base year (2010=100), I will predict the theoretical equilibrium for exchange rate and obtain the misalignments respect the original value. Point out that they are all countries of the European Union, but Sweden and Denmark have their own currencies. Figure 3. RER Each country will be analysed individually using time series, the process of obtaining the misalignments will be explained in the corresponding section 3, when we see the variables used and the econometric process. Secondly, using the results previously obtained I will make an equation of growth including the misalignments calculated in the previous step to obtain the effect in the economic growth in each country. After through impulse reaction we will analyse the comportment of economic growth respect RER misalignments; concluding with a comparative between the north and the south. Summarizing, the objective is to explain how misalignments in the exchange rate affect in the countries of the North and the South. On the other hand, see how the exchange 6 rates are important nowadays for the economic activity using countries of the European Union as the best example to describe this importance. Where the euro area (formed by nineteen countries) is the most important area in the European Union for maintain the stability, nowadays Spain, Greece, Italy and Finland are members of the eurozone. Denmark and Sweden, have they own currencies, but Denmark have linked his currency to the euro with the ERM II (European Exchange Rate Mechanisms II) where the exchange rate of a non-euro area Member State is fixed against the euro and is only allowed to fluctuate within set limits. On the other hand, Sweden still out of this mechanism because the population did not approve it by referendum. The rest is organized first with a review of the literature used to the paper in section 2. In the section 3 we analyse the RER equilibrium equation analysing the components and the results of misalignments. The section 4 use the misalignments obtained in section 3 to create an equation of economic growth for later analyse the coefficient and do an impulse response, to know how growth respond to misalignments in each county. Following in the section 5 are the general conclusions and in section 6 and 7 appendix and data appendix. 2. Literature review There are many studies that relate exchange rate misalignments (defined as deviations of the exchange rate from the equilibrium level) and economic growth, where the use of cointegration models are common. The most differences between the studies are the variables. One of the studies bases use the purchasing power parity from Rodrik (2008) where undervaluation have a good effect over the growth, but in the long-run this suppose does not hold. Other group use the long-run relationship to obtain the misalignment using cointegration time series o panel data, based on a model for determining the exchange rate. Aguirre and Calderón (2006) is an example, the study of the effects of the misalignments in the real exchange rate (RER) for 60 countries during 1965-2003 using cointegration methods for time series and panel. Based on the model of Obstfeld and Rogoff´s (1995) where exchange rate equilibrium is productivity, net foreign assets, the terms of trade and government spending. Concluding that depreciation have a positive effect to the growth, like Rodrik (2008). Similar, Razin and Collins (1999) pose a different model concluding that overvalued currencies have a negative effect to the growth. 7 Other studies are based in the search of which variables include as fundamentals. Berg and Miao (2010) compare the results between the “Washington Consensus” who argues that RER misalignments imply imbalances and in consequence bad for the growth. Although, Rodrick (2008) relate the undervaluation relative to purchasing power parity is good for growth. This study concludes the theory of Rodrik but the viewpoint of WC is more difficult to confirm. Comunale (2017) obtain the same results with the analysis for the EU countries, in the same way Habib et al. (2017) for a large panel of almost 150 countries. On the other hand, Schröder (2013) and Aguirre and Calderon (2005) conclude with an inverse result. The undervaluation is not positive for the economic growth of the countries in comparison with the other studies. Looking around the literature, we can see how misalignments affects growth, undervaluation stimulate growth in contrast than overvaluation that harm the economy. For this, the importance of search asymmetric effects gained importance due to policy. Rodrick (2008), Berg and Miao (2010), Comunale (2017) barely find asymmetries with undervaluation and overvaluation, but Aguirre and Calderon (2006) and Schröder (2013) concludes that both affect negatively to the economy growth where overvaluation with the strongest effect. Other of the studies, Cuestas, Mourelle and Reges (2019) obtain the same results as Berg and Miao (2010), Schröder (2013), Comunale (2017) where overvalued exchange rate is bad for the economic activity for a group of CEE countries. Besides, overvaluation is much stronger than undervaluation. 3. RER Misalignments First, as Cuestas, Mourelle and Reges (2019) do in their paper, to obtain the RER misalignments we selected a group of variables to create an equilibrium equation for the RER. The variables are GDP per capita (pibpc), balance of payments (bp), government expenditure (gov), investment (inv), bond yields (int) and consumer price index (pc). This variable is selected based to literature previously explained and data available, see the data appendix for a definition and the data sources of the variables. Obtained the variables, the equilibrium equation is: 8 rert = α0 + α1 pibpct + α2 bpt + α3 govt + α4 invt + α5 intt + α6 pct (1) This equation is analysed for each country separately, as we know that Purchasing Power Parity does not hold. The RER is proposed as a cointegration relationship of the variables previously exposed to obtain the misalignment. Applying the cointegration test by Johansen (1988, 1991) for each equation the results show a clear relationship in the long-run, there is cointegration in the variables. To see the results of the test look at the appendix. Next, applying the VEC model for cointegrated and non-stationary I obtain the coefficients of the cointegrated model for each country. There are values omitted due to difficult to find all the data for the specific country and year. The estimation in the Table 1 follows the strategy of estimate a long-run exchange rate for time series since 1995 until 2018 in quarters. The coefficient of exchange rate is normalized to 1 and the level of significance is indicated with *. Analysing the Table 1 for each coefficient, first we can see that GDP per capita is significant in all countries but the sign is not the same, with four countries in positive and two in negative. In this case the results are ambiguous, the sign expected is the negative like Greece and Denmark because according to the Balassa-Samuelson effect when a country is more developed should have a more valued currency. This effect relates the differences into a country between tradable and non-tradable market, where the “Penn Effect” says that RER follow the same direction: If the incomes are high, the prices levels are high comparing to international average, and when are low the contrary. According to Cuestas, Mourelle and Reges (2019): “The Balassa-Samuelson effect is the real appreciation generated by the increase in the relative price of non-tradable goods that follows an increase in productivity in the more competitive tradable market. This is driven by the upward pressure on wages in the non-tradable sector that arises because wages in the tradable sector are higher since productivity growth is faster in that sector than in the non-tradable sector”, so in the case of Spain, Italy, Sweden and Finland may be for the no increment in the relative price o and a lower incomes and price levels compared to international average. The next coefficient to analyse it is only available for Spain, Sweden and Finland due to the lack of data in the other year in respective quarters. The balance of payments alone does not serve to explain fluctuations in the real exchange rate, it is composed of a 9 current account and capital account. The balance of payments registers all monetary transactions between a country and the rest of the world, the current account includes net transactions of goods and services, while capital account the inflows and outflows of capital. The sum of these two has to add cero, if current account has surplus the capital account has deficit and vice versa. Continuing with the main part, if the current account has deficit during a long time the currency tends to depreciate. The sing of the coefficients is the expected in Spain and Finland where an improvement in the balance of payments tend to appreciate the currency but in Sweden the results are contradictory may be due to that historically Sweden always has surplus. The government expenditure and investment are all significative in all the cases except the government expenditure in Italy. The interpretation of the signs depends the policy of each country, when a country spends more in non-tradable goods the sign is negative. As we can see Sweden and Finland have a negative sign because the specific policies where spend a lot of money in non-tradable goods. On the other hand, Spain, Greece and Denmark have spent more in tradable goods so for this the sign is positive. With the exception of Denmark that is a interesting difference between northern and southern countries. The explanation for investment is similar, a positive sign indicates that the invest depends more on non-tradable goods and positive if depends more on tradable goods. In this case four countries have positive sign and two negatives, with the exception of Spain and Italy (due to government expenditure is not significative) is surprising that in the other countries government expenditure and investment goes to the same direction. To tradable goods in Greece and Denmark and non-tradable goods in Sweden and Finland. Finally, the interpretation of the last two coefficients are confusing and depend of the characteristics of each country. The coefficients of bond yields are significance with the exception of Spain and oscillate between negative o positive depend the country. The consumer price index has a little significance and it is only significative in four countries, where the signs are positive for two countries and negative for two too. 16 Table 2. VEC of growth. Notes: Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. 17 After applying VEC model I obtain the coefficients for analyse it. The estimation in the Table 2 follows the strategy of estimate a long-run growth for time series since 1998 until 2018 in quarters. The coefficient of growth is normalized to 1 and the level of significance is indicated with *. The first coefficient government expenditure, is only significant for Spain with a negative sign that significate when more spend more growth. It is strange that only for one country is significative but that is not the main component to analyse. Following with investment, this is significative for four countries. Spain Italy Sweden and Finland, the sign is positive with the exception of Spain. In this case the expected sing is positive because when more invest more economic growth, but in this case Italy, Sweden and Finland have a contrary effect with a reduction of growth when investment increase. Leaving the misalignments for later, employments is significative for 5 countries. The sign is positive in three of them and negative in two, it is curious in Spain and Italy because affect negatively in the growth may be for the big impact of the crisis the result is not the expected. Although, in Greece, Swede and Finland the effect is the expected where an augment of employment is related positively to growth. The next coefficient, consumption of households is only significant for the southern countries. In Spain and Italy an augment of consumption increments the growth while in Greece reduces it, so it is a strange comportment of the variable that will be negative. As we can see the sign of coefficients in some cases are not the expected, it does not matter to much because the analysed coefficients of variables are only a complement to create a better equation of economic growth with a misalignment as main component. Before explain the comportment of growth when misalignment vary using impulse response function of VEC models, lets watch the coefficients and the sign of cointegration equation. For all the countries the coefficient is significant, the sign is negative in Italy, Greece, Denmark and Finland meaning a positive relationship between misalignments and growth. On the other hand, Spain and Sweden have a negative relation between misalignments and growth, an increment of misalignments reduces the growth. The sign of coefficient may be explained how is the growth comportment but, in some cases, can not be true this relationship. For obtain a more accurate response of the Growth when misalignments changes, we use the Impulse Response to Cholesky One S.D. Innovations. Through this it is possible obtain in a better way the effect on the 18 growth. After all, using the VEC model it is possible obtain the long-run relationship of misalignments with the growth, in this case for a period of 40 quarters. The first country to analyse is Spain, once obtained the Figure 9 there is a clear trend. The effect in the growth is positive, during the first periods the influence of misalignments onto growth grow up from 0% in the period 1 to 0.06% in the period 3 when the influence stabilizes until period 5. Starting to this period the influence starts to decrease period after period until period 12 that we can see a change, after that the influence decrease more until -0.04% when stabilizes. Anyways, for a long period of ten years the pattern, even though is positive at the beginning, is negative misalignments affect negative to growth as the sign of coefficient. Figure 11. Spain impulse response The next country is Italy, in the Figure 10 we can see the results of the response of grow to misalignments. Starting from 0% in the first period the relation goes down until -0.09% approximately in period 2, since this period the trend changed. After period 2 the effect of misalignments to growth recover the initial value 0%, but after this the relationship starts to be positive. As the highest point in the period 7 with almost 0.08%, after some positives periods but with a big variance the value stabilizes around 0.03%. In this case, as the sign of the coefficient, the relationship between the growth and misalignments is positive, even though has a few periods of negative relation. 19 Figure 12. Italy impulse response Following with the last southern country, Greece. The results are in the Figure 11. In this case the relation between growth and misalignments is always positive, highlight the period 23 when the value is almost 0.12%. After this maximum the value fall to 0.02% and start to oscillate around 0.06% and 0.04%, for stabilizing in the last periods around 0.05%. In the same form as Spain and Italy the response of growth to misalignments is like the sign of the coefficient obtained previously in VEC model. Figure 13. Greece impulse response 20 Starting with the first northern country, Denmark. The results are in the Figure 12. That results are similar to Greece but with less oscillation, at the beginning the value grow bit a bit the first 2-3 periods for jump to 0.12% at period 4. After, the value falls to 0.06% in period 6 for in the following periods oscillate between 0.11 and 0.09, and stabilize around 0.10% in the last 20 periods. As que can se the relation in this case is positive all the time with a big impact on the growth, following the same sign obtained in the coefficient. Figure 14. Denmark impulse response Sweden has a curios comportment, form the first period fall to -0.06% in period 2 to grow up to 0.03% in period 3 and for finally fall to -0.08% in period 4. After this the value stabilizes in -0.08% some periods and fall with a little oscillation to -0.10%. In this case the relation is almost all the time negative, where misalignments affect negatively to economic growth. As the other countries the coefficient of VEC model coincide with the prediction of ten years even having a period of positive relation. 21 Figure 15. Sweden impulse response The last is Finland. In the Figure 14 are the results. As Sweden, Finland has a strange comportment the first periods. In the periods between 2-3 the value is almost 0.12% for after fall to -0.01% in period 4, the next periods the oscillation is big around -0.01% and -0.04%. In this case the coefficient of VEC model it does match with the results, may be due to the big positive value that we can see in the period 3, anyways the conclusion is that misalignments have a negative effect to the economic growth. Figure 16. Finland impulse response 22 Once the countries are analysed, we can get some conclusions. The equation to relate economic growth and the misalignments has worked fine to obtain a consistence result. Although, in the VEC coefficients some of the variables have no signification the misalignments are significant in all the countries, where the sign with the exception of Finland are in the same way as the impulse response. Respect the north and the south, in the two cases are one exception, Italy and Greece have a positive effect from misalignments to the economic wroth while Spain has a negative. Sweden and Finland have both a negative effect from misalignments to economic growth while Denmark has a positive. Using the information obtained in the section 3 where we calculate, through cointegration relationship, the theatrical equilibrium for each country, that we have obtained of the coefficients of the equation for RER. It is possible confirm one of the results previously saw in the literature. In the graph comparing the RER and que theoretical equilibrium from Sweden we see that the currency is undervalued, linking this to response impulse analysis can conclude that undervaluation has a negative effect to the economy. As Aguirre and Calderon (2006) and Schröder (2013) concludes, that over-valuated and undervalued affect negatively to the economy growth. On the other hand, looking the graph of Italy can see an asymmetry because in the graph the tendency of Italian currency is over-valuated while in the growth analysis in the case of Italy the misalignments affect positively. So, the theory of Aguirre and Calderon (2006) and Schröder (2013) does not hold in this case. 5. Conclusions The objective of this project is understanding how exchange rate are important for the economy and the possible effects. In a world where the exchange rates are increasingly important, with the clear example of China and it is policy of undervaluation for gain competitiveness. This added to the Balassa-Samuelson effect were variations of exchange rate may have important consequences to the economy. Selecting three northern countries (Denmark, Sweden and Finland) and three southern (Spain, Italy and Greece), with the simple reason to compare two realities inside the EU as euro area as a reference due to euro is the second most important currency at the world. The misalignments are the main component of the project, through an equilibrium equation for RER. Applying cointegration by VEC model we obtain some results that 23 using the coefficients of cointegration can compute the theoretical RER equilibrium for each country. Spain before the crisis had an undervalued exchange rate but after the crisis the misalignments had reduced and the last years of recovery form the crisis the exchange rate started to stay over-valuated. Italy had an over-valuated exchange rate during all the analysed period and Greece has a similar result to Spain but with a high variance and the last years the exchange rate started to stay undervalued. Denmark is divided into three periods, the first the currency is undervalued, in the second the tend change and is over-valuated and finally in the last the currency is starting to stay undervalued. In the Sweden case the currency is clearly undervalued during all the period analysed. Finally, Finland is hard to analyse because a high variance during all the periods that makes impossible analyse the state of exchange rate. The results of this is much more misalignments in the northern countries than southern with the exception of Denmark, with Finland and Sweden on the head followed by Greece, Italy, Spain and Denmark respectively. So, we can see the first differences between north and south. These misalignments are crucial because we include it into the next regression of growth, the process is the same as previously, highlight the big significance of the coefficient misalignments in all the countries. Through impulse response in the VEC model we obtain the last results. In Spain the relation of misalignments and growth is positive the first periods but later this begins to be negative little by little. Italy with a negative peak at the firsts periods becomes positive with the following periods with a big positive peak for later stabilize to a lower value. Greece characterized for at a big positive peak at the beginning for stabilizes to a lower positive value in the next periods after some variance. Denmark similar to Greece peak at the beginning and stabilizes in the next periods but a high value. Sweden has a strange behaviour, first the relation is negative with a peak in the next to periods the relation is positive with a big value and finally fall to a negative value bigger than the first peak. Finally, Finland start with a big positive value and fall to a negative in the next periods where vary between 0.01% and 0.04%. In the north and the south, we see a tendency with an exception in both. Sweden and Finland have a negative effect of misalignments in the economic growth while Denmark has a positive relation being the biggest respect the positive countries. Italy and Greece have both positive effect of misalignments over the economic growth but Spain has a negative relation. It is hard getting a conclusion of this results because we would need more countries for analyse the behaviour for discard that the case of Spain and Denmark 24 are isolated cases for conclude. In the case of Denmark, the fact that his currency is linked to the euro with ERM II, that does not allow fluctuate more than 15% from the euro can have the explanation of the different result respect the northern countries. In the case of excluding Spain and Denmark, the differences between the north the south are clear. The northern countries have a negative relation between misalignments and growth and the southern have a positive relation between growth and south, other pattern is that Sweden and Finland have more high misalignments compared to Italy and Greece. So, in this theoretical case we can conclude that if there are to much misalignments probably the effect of this in the economy will be bad, as the case as Sweden and Finland. Mixing the results of the RER and growth sections, we can see a relation saw in the literature. If we compare the theoretical RER equilibrium of Sweden with the conclusion of all the time the currency is undervalued linked to the conclusion of negative effect of misalignments into the growth. We can prove on part of the theory of Aguirre and Calderon (2006) and Schröder (2013), that over-valuated and undervalued affect negatively to the economy growth. Although, looking the case of Italy we have an invers result respect Aguirre and Calderon (2006) and Schröder (2013), in this case the exchange rate of Italy is overvalued and the effect of misalignments is positive so the theory does not hold. Highlight that this mix of the two sections was done with these countries because in the theoretical exchange rate have a clear tendency. Anyways, that theory does not hold, one thing is clear in this project. Exchange rates play a crucial role in the growth of countries and more in an increasingly globalized world, whether negative or positive these misalignments cause as we have seen reactions in countries and their growth for this the importance of the policies will increase more. In this work we have not been able to obtain a clear answer about the possible differences between the countries of the north and the south, because of the lack of analysing more countries, or through a more select selection of variables that would allow us to get closer to reality. Although there is an individual response for each country and the effects that the misalignments have on them. 25 6. Appendix In the first moment in the project we want to use a VAR model and apply an impulse response of the variables because is a perfect model for a time series. But for use this model we need stationary variables, hard thing to get. Continuing, checking the stationarity, all the variables are integrated ergo no stationary. After the first analysis we can not use the VAR model, in this case que can do two things. Transform the variables in stationary, for example using growth rates of the variables but if we do this the possible relations of cointegration are eliminated and the we lose important information. Or using the cointegration test by Johansen (1988, 1991) check if the variables have a long-run relation for use a Vector Error Correction (VEC) model, that is an extension of VAR model for cointegrated and integrated series. We use the cointegration test by Johansen (1988, 1991) in the RER equation from section 3. In the Table 3 we have the test for Spain, Italy and Greece respectively, as we can see the p-value of MacKinnon-Huag-Michelis at 0.05 level indicate at least four cointegration equations for Spain and Italy and three for Greece. As the Table 3, Table 4 shows the results for Denmark, Sweden and Finland. Where the p-value at 0.05 we can see Denmark with three cointegration equations, Sweden with two cointegrated equations and Finland with seven cointegrated equations.