1
Fire-sale Foreign Direct Investment during the euro-crisis:
Did foreign investors take advantage of the crisis in PIIGS?
Victor G.A. Munnichs
*Master Thesis
Amsterdam, June 20
th2014
Supervisor: Dr. H. Vrolijk Co-assessor: Dr. C. Hermes
University of Groningen Faculty of Economics and Business MSc International Financial Management
University of Uppsala Faculty of Social Sciences MSc Business and Economics
*
Student-number: s1774336 Cornelis Anthoniszstraat 32 III 1071VV Amsterdam
2
Fire-sale Foreign Direct Investment during the euro-crisis:
Did foreign investors take advantage of the crisis in PIIGS?
ABSTRACT
During previous crises, foreign investors increased their foreign direct investment (FDI) activity in the countries which suffered of the crisis. They acquired firms with liquidity problems at fire-sale prices. In this paper I investigate if foreign investors acted the same during the euro-crisis in weak-performing Eurozone countries. These weak performing countries are the PIIGS, an acronym for Portugal, Italy, Ireland, Greece and Spain. I investigate data on inward FDI flows/stocks and data on completed acquisitions. My results show that during the euro-crisis foreign investors relatively increased their investments in the PIIGS-countries. As first in literature, I investigate if the increase in foreign investments comes from regions which did not suffer of the crisis. I find significant results that firms of the U.S. and non-European regions relatively increased their investment activity in the PIIGS while firms of other-European countries did not. Thereafter, I found that besides foreign financial firms also foreign non-financial firms relatively increased their acquisition activity in PIIGS. As a result, I expect that the increase in foreign investments was not only due to short-term financial motives. Finally, the finding that foreign investors increased the amount of controlling stake acquired is in favor of the fire-sale FDI hypothesis.
3 Table of content
1. Introduction ... 4
2. Theoretical Background ... 5
2.1 Fire Sale FDI ... 5
2.1.1 Evidence for the increase in inward FDI ... 5
2.1.2 Reasons of the increase in inward FDI ... 6
2.1.3 Acquisition of stake and control ... 7
2.1.4 Flipping the acquisition ... 8
2.2 Overview of the hypotheses ... 9
3. Data ... 10
3.1 Data on Foreign Direct Investments ... 10
3.2 Data on Mergers and Acquisitions ... 11
3.3 Two types of Data ... 12
4. Results 1 – FDI data ... 12
4.1 Inward FDI in the PIIGS countries ... 12
4.1.1 Flows ... 12
4.1.2 Stocks ... 13
4.2 Inward FDI in the PIIGS countries per region of origin ... 14
4.2.1 Stocks ... 14
4.2.2 Flows ... 15
5. Results 2 – Acquisitions data ... 16
5.1 Features of the acquisitions data ... 16
5.1.1 Acquisitions in PIIGS ... 16
5.1.2 Cross-border M&A in PIIGS ... 17
5.1.3 Financial vs. Non-financial ... 18
5.2 Regression Analyses – Per Region ... 19
5.2.1 Methodology ... 19
5.2.2 Results ... 20
5.3 Regression Analyses – Financial vs. Non-Financial Sector ... 22
5.3.1 Methodology ... 22
5.3.2 Results ... 22
5.4 Regression Analyses - Degree of Corporate Control Sought ... 23
4
1. Introduction
As in 2009 the European Commission celebrated the tenth anniversary of the euro, she analyzed that the Eurozone had been relatively unscathed by the 2008 global financial crisis (Gibson et al, 2009). This view changed dramatically in the beginning of 2010 when the revelation of Greeks’ major sovereign debt problems led to a snowball effect. The problems spread through the euro-zone as other euro-countries like Ireland and Portugal suffered of sovereign debt problems as well. This led to a crisis called the “euro-crisis” (Gibson et al, 2014; Gorea & Radev, 2014; Polita & Wickens, 2014). During this euro-crisis the most troubled countries became called the ‘PIIGS’, an acronym for Portugal, Ireland, Italy, Greece and Spain (Mundell, 2012).
After the 1997 Asian crisis, newspaper headlines on the sale of assets at dump prices in countries hit by the crisis were the reason for Krugman (2000) to come with a theory called “fire-sale Foreign Direct Investment (FDI)”. Fire-sale FDI takes place when decreasing domestic asset prices together with an access to financial capital by foreign investors leads to the sale of domestic assets to foreign firms at bargain prices (Poulsen & Hufbauer, 2011). This concept is often used to research the pattern of foreign direct investment during the Asian crisis in 1997 and to less extent the crisis in Latin America in 1995. In this study I attempt to link the euro-crisis in the PIIGS-countries to Krugman’s (2000) theory of “fire-sale FDI”. The theory of fire-sale FDI has not been investigated for the recent crisis yet.
Calderon and Didier (2009) mention that fire-sale FDI was not of impact during the 2008 financial crisis because it was a global crisis of which countries in all regions suffered. However, in contrast to the financial crisis the euro-crisis occurred in the Eurozone only. Moreover, the sale of assets to foreign investors at dump prices in these weakened countries is often mentioned in newspaper articles such as “Portugal to hold fire-sale of state assets” (The Guardian, 2012); “Israeli firms compete in Greece fire sale” (The Jerusalem Post, 2012) and “Everything Must Go! The Great European Fire Sale” (The Independent, 2012).
I will investigate changes in the pattern of inward foreign direct investments in PIIGS during the euro-crisis compared with the pre-euro-crisis period. As one of the first, to my knowledge, I will focus on differences in crisis-time inward direct investments coming from different regions. The study is built on answering the main question: “Have there been changes in the pattern of inward foreign direct investment in PIIGS during the euro-crisis and is this in line with the theory of fire-sale FDI?”
5
value of inward FDI flows and stocks, as well as the number of inward foreign acquisitions. The following chapter will discuss these results and the last chapter concludes.
2. Theoretical Background
2.1 Fire Sale FDI
There is not much research on the relation between foreign direct investment (FDI) and crisis. Krugman (2000) was one of the first who came up with the concept of “fire-sale FDI”. He questions “why direct investment surge at a time when foreign capital in general is fleeing a country” (p. 44). Subsequently he wonders if this is the result of the special competences of the acquired firms or simply because foreign firms have cash while domestic firms do not. The concept of fire-sale FDI of Krugman is specifically based on the Asian crisis in the late 1990s and, less marked, on the crisis in Latin America in 1995. His theory builds on the financial panic point of view. This view is based on the assumption that domestic firms have severe liquidity problems during a crisis period in comparison with the foreign investors. Earlier work of Schleifer and Vishny (1992) describes the fire-sale of assets of firms with liquidity problems. Krugman extends this work and states that even though the foreign investors are less efficient in running the firm, they are still more willing to acquire the firm than the domestic investors because they have access to liquidity.
While Krugman did not empirically investigate fire-sale FDI, subsequent papers did in-depth research on this subject. Aguiar and Gopinath (2005) were one of the first who empirically investigated the topic for the Asian crisis. Subsequently, Acharya et al (2011) extended the literature by providing theory on the level of control sought and the concept of “flipping the acquisition”. Recently, Alquist et al (2013) investigated fire-sale FDI for several crises and provided a benchmark article for this thesis. The main topics discussed in the previous literature on fire-sale FDI will be discussed in the coming sub-paragraphs.
2.1.1 Evidence for the increase in inward FDI
To provide empirical evidence for fire-sale FDI during the 1997 Asian crisis, Aguiar and Gopinath (2005) first looked at the magnitude of inward FDI flows in the countries hit by the crisis. In contrast with a significant decrease of other inward investments2, the magnitude of inward FDI flows slightly increased during the crisis. Besides this, they also found that the number of inward foreign mergers and acquisitions in crisis countries increased between 1996 and 1998 by 91% while domestic M&A decreased by 27%. In addition to Aguiar and Gopinath (2005), Alquist et al (2013) researched several financial crises and found
6
that the likelihood of a firm being acquired by a foreign company during a banking crisis is higher than in non-crisis periods. Although the crisis-effect is highest for Asian countries during the 1997 crisis it also holds for the other crises investigated (i.e. the crisis in Latin America and South Africa).
Acharya et. al (2011) found that in Asia during the crisis, besides the relative number of acquisitions, the relative total deal value of foreign transactions increased significantly. Total deal value of foreign transactions as percentage of total value of all transactions grew of 14.3% between 1991-1994 to 39.7% between July 1997 and December1999. They focused on the relation between inward FDI and foreign portfolio investment (FPI) during the Asian crisis. Foreign portfolio investments are short-term foreign investments in small shares, not for control or long-term improvements but for diversification of the investment portfolio. In non-crisis times, FDI and FPI were tested to be positively correlated. However, during the crisis in Asia, FPI dried-up almost completely while FDI rose significantly.
In contrast with these findings in favor of the fire-sale FDI theory of Krugman, Poulsen and Hufbauer (2011) describe that even though the numbers of the cross-border transactions in the crisis hit countries increased during the Asian crisis, these increases were limited if compared with the numbers in other regions (i.e. Brazil and Argentina). Besides, the total value of deals in Asia were limited if compared with the value in other regions. They ascribe the increase of inward FDI in Asia to policy intervention by the governments instead of the fire-sale prices of assets.
Based on the findings of Aguiar and Gopinath (2005), Acharya et al (2011) and Alquist et al (2013) I expect that the number of inward foreign mergers and acquisitions in PIIGS relatively increased during the euro-crisis. Therefore my first hypothesis states the following:
Hypothesis 1: The number of inward foreign acquisitions in PIIGS increased during the euro-crisis relatively to the total number of acquisitions in these countries.
2.1.2 Reasons of the increase in inward FDI
The literature on fire-sale FDI is mainly focused on the cheap assets hypothesis, also highlighted by Krugman. The cheap assets hypothesis is based on the idea that the targets assets are priced below their initial benchmark price due to a collapse of investor sentiment or a stock market crash and therefore are cheap opportunities for foreign investors (Baker et al, 2009).
7
being acquired. The “decline in firm liquidity between 1996 and 1998 can account for 25% of the observed increase in acquisition activity” (p.440; Aguiar and Gopinath, 2005).
During the Asian crisis the price paid for an acquisition fell from a price-to-book ratio of 3.5 in 1996 to 1.3 in 1998 due to an increase in the acquisitions of liquidity constraint firms (Aguiar and Gopinath, 2005). The liquidity constraint firms were relatively more often targets for foreign acquirers than for domestic acquirers. Aguiar and Gopinath state that foreign acquirers buy the domestic acquisitions because these cannot finance themselves since they are liquidity constraint. Alquist et al (2013) find that firms in industries which are more dependent on external finance are more likely to experience acquisitions of foreign investors. However, this does not differ for crisis and non-crisis periods since the crisis coefficient is insignificant.
Alquist et al (2013) expect that especially foreign firms of the financial sector take advantage of the liquidity problems of the firms in crisis countries. It is hypothesized that they buy the liquidity constraint firms at dump prices to improve them for short term goals such as fast increase in cash returns and to resell the investment at a higher price after the crisis (see paragraph 2.1.4). However, against their expectations they found that during crises the probability that an acquisition had a foreign non-financial acquirer increased more than the probability that an acquisition had a foreign financial acquirer. Besides, they found that the probability of acquisition increased during crises for foreign firms acquiring non-financial targets while this was not significant for foreign firms acquiring non-financial targets. This was against their expectation that mainly financial targets would be acquired relatively more during crises. Although Alquist et al (2013) did not find significant evidence for their statement that foreign financial firms take advantage of the fire-sale priced assets; I concur with them and state that acquisitions of foreign financial firms will increase during the euro-crisis.
Hypothesis 2: The probability that an acquisition had a foreign financial acquirer increased during the euro-crisis more than the probability that an acquisition had a foreign non-financial acquirer (relatively to total acquisitions in PIIGS).
2.1.3 Acquisition of stake and control
8
before the crisis by both foreign and domestic acquirers. This implies that during a crisis it is more likely that investors receive the majority control due to the acquisition.
Alquist et al (2013) distinguish results between the probability of control sought and the amount of shares acquired during crisis. They do not find significant results that the patterns of acquiring control differ between crisis and non-crisis periods. However, they found significant results when testing if the amount of stake acquired differs. Against expectations, the results show that during times of crisis foreign investors acquire a smaller amount of stake than in non-crisis periods. This finding contradicts the “fire-sale” FDI theory that foreigners acquire larger stakes during a crisis.
As a result of the finding of Acharya et al (2011) I expect that during the euro-crisis foreign investors may find it more profitable to acquire control. Especially foreign financial investors, which are expected to take advantage of the fire-sale priced assets, should increase its amount of control sought. For this expectation I formulate the following hypothesis:
Hypothesis 3: The control sought by foreign acquirers, especially by foreign financial acquirers, in PIIGS increased during the euro-crisis if compared with the pre-crisis period.
2.1.4 Flipping the acquisition
An additional feature of the data on fire-sale FDI is the concept of “flipping the deal”. Acharya et al (2013) mention that the acquisitions at fire-sale prices made during a crisis are expected to be for short-run motives rather than for long-short-run investment plans. Financial investors can buy a target at a fire-sale price, restructure and improve the company and when the crisis disappears sell back to the domestic investor. Therefore it is expected that the acquirer-target relationships of crisis time acquisitions are of shorter length.
They test “how many of the new owners do, indeed, sell and exit relative to the cyclical benchmark given by the level of re-selling by domestic investors” (p.3). It is found that foreign investors are more likely to flip their investment. Foreign Acquisitions made during the Asian crisis between July 1997 and December 1999 are, if compared to domestic acquisitions during the same period, almost twice as likely flipped within a period of five years. In the pre-crisis period, foreign acquisitions are not more or less flipped within a five year period than domestic acquisitions. Flipping rates seem to be higher in the countries which are hit very hard by the crisis if compared to the whole sample. Moreover, flipping rates in countries mildly hit by the crisis do not significantly differ of non-crisis time rates.
9
between crisis and non-crisis periods while domestic acquisitions face lower flipping rates during crises compared with non-crisis periods.
In addition, they investigate if the acquirers’ and targets’ industries are of influence on the likelihood of being flipped. It is found that transactions between two firms in the same industry have less chance of being flipped. Financial acquirers, domestic and foreign, have higher flipping rates than non-financial acquirers. This is in line with the expectations of Acharya et al (2011) that financial investors seem to buy firms during fire-sale to resell later for higher prices.
To test if inward foreign acquisitions in PIIGs during the euro-crisis were flipped more often than pre-crisis foreign acquisitions, data is needed on acquisitions made after the pre-crisis. As the euro-pre-crisis only recently has occurred, no data on flipping rates is available yet. Therefore no direct research can be done on the concept of “flipping the acquisition” in this thesis. However, as mentioned in paragraph 2.1.2, flipping the acquisition is expected to be a motive for especially foreign financial acquirers when they take advantage of assets at fire-sale prices. Therefore, the results according to hypothesis 2 (if foreign financial firms acquire more during crisis) can be used to make predictions on the flipping rates.
2.2 Overview of the hypotheses
Table 5.1 shows an overview of the previously stated hypotheses and the theoretical evidence of the three main articles (Aguiar & Gopinath, 2005; Acharya et al, 2011; Alquist et al, 2013). As can been seen an extra hypothesis is added on the region of origin of the foreign investor (hypothesis 4). The literature states that the increase in FDI into crisis-countries is a result of domestic firms having severe liquidity problems during a crisis period of which foreign investors take advantage (Aguiar and Gopinath, 2005). Although not investigated in previous literature (to my knowledge), it can be stated that foreign investments should increase from regions which do not suffer of the same crisis, otherwise the foreign firms suffer from liquidity problems as well. Thus, during the Euro-crisis, it is expected that other Eurozone countries did not face an increase in their direct investments into PIIGS while non-Eurozone countries did. This leads to my last hypothesis:
10 Table 2.1 Hypotheses and theoretical evidence
Hypothesis Aguiar &
Gopinath (2005)
Acharya et al (2011)
Alquist et al (2013) H1 Relative increase in acquisitions with a
foreign acquirer Yes Yes Yes
H2 Relative increase in acquisitions with a
foreign financial acquirer n.i. n.i. No
H3 Relative increase in control sought n.i. Yes No
H4 Relative increase in acquisitions with a foreign acquirer of regions which did not suffer of the euro-crisis
n.i. n.i. n.i.
Notes: “Yes” refers to results that are in line with the hypothesis. “No” refers to results that are not in line with the hypothesis. “N.i” refers to “Not investigated”, included in this table since some topics of the hypotheses have not been investigated.
I will not start my research before mentioning that some articles imply that the concept of fire-sale FDI does not fit the recent financial crisis. According to Calderon and Didier (2009) fire-sale FDI was not of impact during the 2008 financial crisis. They mention that this crisis had a global impact, in comparison with the Asian and Latin American crises, and therefore not only affecting the financially constrained firms but also the potential foreign buyers of these firms. Besides, Desbordes and Wei (2012) argue that this existing literature is only applicable to financial crises in emerging economies with relatively underdeveloped financial markets where firms are not able to gain access to credit themselves. The attractiveness of firms in countries attacked by a financial crisis for foreign investors should reduce when the large domestic firms dominate and when the local market is dominantly important for the sales of the company.
The next section will state the data used to test these research questions. Thereafter, chapter 4 and 5 include the methodology and results while the hypotheses will be discussed in chapter 6.
3. Data
3.1 Data on Foreign Direct Investments
11
3.2 Data on Mergers and Acquisitions
For data on the number of mergers and acquisitions in PIIGS I used the SDC Platinum Thomson Reuters database. Greenfields are also part of FDI but not included in this database. This is not a problem because greenfield investments are not relevant for a study on fire-sale FDI, this is due to the fact that a greenfield is not an existing business and therefore cannot be at “fire-sale” price. Since mergers are not included as being “mergers” in the SDC-database, I will refer to M&As as acquisitions throughout the next chapters. The SDC Platinum database has data available on the date of the acquisition, the amount of the acquired shares, the targets’ and acquirers’ firm nationality and the targets’ and acquirers’ SIC-industry classification.
I used the 1st of January 1999 as starting date because of the introduction of the euro. I make a separation between two different crises: the financial crisis and the euro-crisis. I use the 1st of October 2008 as starting date of the financial crisis. The euro-crisis started between December 2009 and February 2010. Therefore I use the 1st of January 2010 as the starting date of this crisis. The included transactions have to be completed deals in which the investor acquires a minimum stake of 5%, to exclude transactions made for portfolio investments. My dataset contains 21,901 acquisitions of which 15,526 domestic acquisitions, 661 intra-PIIGS acquisitions and 5,741 acquisitions with a foreign acquirer3. Table 3.14 shows the data per country of the target and region of the acquirer5.
I added several control variables to my dataset which I use in the regression analyses (Chapter 5). In the selection of the control variables I follow Alquist (et. al., 2013). Data for two of the control variables are obtained of SDC platinum. These control variables are the percentage stake acquired6 and country×industry fixed-effects vectors. Country×industry fixed-effects are included as a set of dummy variables with a dummy variable for each combination of target country and target industry. I also control for macro-economic conditions by using the targets’ country real GDP, real GDP-growth, currency depreciation and use of IMF credit as a percentage of quota. This is done to check for country-level conditions which influence a firms’ decision to invest in a company. This data is gathered of the Worldbank database (real GDP and real GDP-growth) and the IMF international financial statistics database.
3 In Appendix A a list is attached of all acquisitions made by Dutch firms in PIIGS during the euro-crisis. 4
In Appendix B more tables are included on the acquisitions data. In these tables, domestic and foreign acquisitions per target country and per acquirer region are reported for each year of the period 1999-2013.
5 Acquirers are divided in five regions: Domestic, Intra-PIIGS, EU7, Rest of Europe, Rest of the World (ROW) and
U.S. Intra-PIIGS are the other-than-domestic acquisitions by PIIGS countries, EU7 are the original Eurozone countries minus PIIGS, “Rest of Europe” are the other countries in Europe, U.S. are the United States and “Rest of the World” (ROW) are all other countries in the world.
12
Table 3.1 Number of acquisitions for each target country by region of the acquirer Target Country
Acquirer Region Portugal Ireland Italy Greece Spain Total
Domestic 1,039 846 5,278 1,018 7,336 15,526
Intra-PIIGS 178 19 136 28 300 661
EU7 135 116 857 81 842 2,031
Rest of Europe 103 466 568 98 647 1,882
U.S. 35 248 437 24 423 1,167
Rest of the World 72 72 214 31 245 634
Total 1,562 1,767 7499 1,280 9,793 21,901
3.3 Two types of Data
Both datasets have their advantages and disadvantages. Data on the number of acquisitions is relevant to use according to the fire-sale FDI hypothesis since the number of transactions is interesting for the question if the foreigners took advantage to buy domestic assets during the crisis. However, this data does not contain information on the total value of the investments since the value of the transactions is not disclosed for 65.1% of the included acquisitions.
In contrast, data on inward FDI flows and stocks gives information on the value of the inward investments but does not contain information on the number of transactions. A lower value of inward FDI stocks or flows in a country during a crisis does not mean that fire-sale FDI did not happen. The price of the domestic assets can be weak and therefore the flow/stock seems to be decreased. At the same time, the number of transactions of foreign investors can be increased. Then, the investors took advantage to buy the domestic assets at fire-sale prices. Therefore the combination of data on the value of inward FDI and data on the number of inward cross-border mergers and acquisitions gives a complete overview of foreign direct investments in the PIIGS.
4. Results 1 – FDI data
4.1 Inward FDI in the PIIGS countries
4.1.1 Flows
13
To see how the inward FDI flows in PIIGS relate to the inward flows in other countries the figure also shows the PIIGS’ percentage of the total inflows in EU127
, EU278 and the world. As can be seen, the relative position of the PIIGS in 2010 was higher than in the previous years. Relatively more FDI flew to the PIIGS countries, especially as a percentage of EU12 and EU27. Thus, during the Euro-crisis foreign investors invested relatively more in PIIGS than before the crisis. The opposite is true for 2008, at the start of the financial crisis, when a lower percentage of EU12 and EU27 inward FDI flows went to PIIGS. When looking at the position of PIIGs compared to total inward FDI flows in the world, the position of the PIIGS in 2010 only slightly increased.
Figure 4.1, Inward FDI flows in PIIGS, total value and as percentage of total inflows in the World, EU12 and EU279
4.1.2 Stocks
The inward FDI stocks in PIIGS annually increased slightly as can been seen in figure 4.2. At the start of the financial crisis, in 2008, a small breakpoint is noticed because total inward FDI stock decreased by 0.8%. No turning-; or breaking point is visible at the start of the euro-crisis in 2010. Inward FDI stock even increased this year by 7.2%, which is more than the average annual increase of 6.3%. The PIIGS’ inward FDI stock compared with total stock in the world, EU12 and EU27 increased only slightly during the euro-crisis.
7 EU12 are the original euro-zone countries which introduced the euro in 2002. Besides the PIIGS, this group of
countries also exists of Austria, Belgium, France, Germany, Luxembourg and The Netherlands.
8
EU27 are the European Union countries. Countries which belong to EU27 are: all countries of EU12, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia, Slovenia, Sweden and the United Kingdom
9 All reported values in millions of U.S. dollars.
0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 0 20000 40000 60000 80000 100000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
14
Figure 4.2, Inward FDI stocks in PIIGS, total value and as percentage of total inflows in the World, EU12 and EU27
4.2 Inward FDI in the PIIGS countries per region of origin
4.2.1 Stocks
In figure 4.3 the inward FDI stocks in PIIGS of figure 4.2 are divided between five regions of origin.10 Now it can been seen which region has the largest share in the inward FDI stock in PIIGS. Most of the inward FDI comes from EU7 countries. Also the “Rest of Europe” has significant stocks in PIIGS. The stocks of the other PIIGS, the U.S. and the “Rest of the World” (ROW) are small, compared with the other regions.
Besides this, also the pattern of the inward FDI stocks per region is now visible. The stocks of most regions face the same, upward, trend even during the financial crisis; and euro-crisis period. In 2008, at the beginning of the financial crisis, U.S.; EU7 and intra-PIIGS investors reduced their inward FDI stocks in PIIGS by 11.1%, 2.3% and 6.5% respectively. ROW investors only increased its stock slightly by 2.3% while “Rest of Europe” investors increased their inward FDI stocks into PIIGS by 13.0%. Overall, countries attacked by the financial crisis as first, i.e. the U.S. and euro-zone countries, were also the countries which reduced its inward FDI stock in PIIGS during the financial crisis.
In contrast with the reduction in the inward FDI stock in PIIGS of most regions at the beginning of the financial crisis, during the first year of the euro-crisis (non-European) foreign investors increased its stock heavily. “Rest of Europe”; U.S. and ROW investors increased its stock in 2010 with 19.4%, 19.5% and
10 The five regions are: PIIGS, EU7, Rest of Europe, U.S. and ROW. PIIGS are the other-than-domestic PIIGS
countries, EU7 are the original Eurozone countries minus PIIGS, “Rest of Europe” are the other countries in Europe, U.S. are the United States and ROW are all other countries in the world.
0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 0 200000 400000 600000 800000 1000000 1200000 2005 2006 2007 2008 2009 2010 2011 2012
15
15.3% respectively while the stock of EU7 and PIIGS investors changed by 2.3% and -0.7% respectively. This is in line with the fire-sale hypothesis that countries which did not suffer of the crisis, i.e. the non-euro countries during the non-euro-crisis, increase its investments in the countries hit.
Figure 4.3, Inward FDI stocks PIIGS, total value per region
4.2.2 Flows
Also the inward FDI flows can be separated per region. This is visible in figure 4.4. As can been seen, the flows per region fluctuate heavily over time. The largest FDI inflows in PIIGS come from other European countries. Intra-PIIGS FDI inflows were low, except in 2007. The only negative intra-PIIGS FDI inflow was in 2010 at the start of the euro-crisis. The flow from EU7 countries is very high in half of the reported years but much lower in 2008 and 2010. During these years a crisis started in the euro-countries (the financial crisis in 2008 and the euro-crisis in 2010). The fact that EU7 and PIIGS countries suffered during this crisis can explain the lower inward FDI flows to PIIGS.
In contrast, the 2010 inward FDI flows in PIIGS of the U.S., “Rest of Europe” and ROW were higher than in most other years. The euro-crisis did not lead to a decrease of the inward flows which can be evidence for the fire-sale hypothesis. The inward FDI flows of the United States were negative in 2008 and 2009. During these years the U.S. suffered from the financial crisis. In 2010 the inward FDI flow in PIIGS was positive again and would remain positive during the subsequent years. It is striking that the inward flows of the rest of the world were highest in the crisis years 2009, 2010 and 2012.
0 100000 200000 300000 400000 500000 600000 700000 2005 2006 2007 2008 2009 2010 2011 2012
16
Figure 4.4, Inward FDI flows PIIGS, total value per region
5. Results 2 – Acquisitions data
5.1 Features of the acquisitions data
The data on acquisitions provides me with direct information on the (relative) pattern of foreign acquisitions in PIIGS in the pre-crisis and crisis period. In the results that follow in this chapter “foreign acquisitions” are acquisitions in PIIGS made by foreign acquirers. Intra-PIIGS acquisitions are not included under “foreign acquisitions”. Direct information is available on the sector of origin of the foreign acquirer. I divided the sectors into two groups in financial sectors and non-financial sectors. In the rest of this thesis, a financial firm is defined as a firm of the “finance, insurance and real estate” (FIRE) sector and a non-financial firm as a firm of any other than FIRE-sector. The direct features of the dataset will be ascribed in this section before the data will be statistically tested in the subsequent sections.
5.1.1 Acquisitions in PIIGS
Figure 5.1 and 5.2 show the total number of acquisitions (domestic and foreign) per PIIGS-country. The dashed line in the figures represents the start of the financial crisis in 2008 and the continuous line the start of the euro-crisis in 2010. In most countries the number of acquisitions decreased during both crisis compared with the pre-crisis period. Exceptions are the increase in the number of acquisitions in Greece during 2009 and 2010, Portugal in 2010 and the increase in Ireland during 2012.
-20000 -10000 0 10000 20000 30000 40000 50000 2005 2006 2007 2008 2009 2010 2011 2012
17
Figure 5.1, Total no. of acquisitions in Portugal, Ireland, Greece Figure 5.2, Total no. of acquisitions in Spain and Italy
5.1.2 Cross-border M&A in PIIGS
According to the fire-sale FDI theory it is expected that the number of foreign acquisitions will increase relatively to total acquisitions. The difference between foreign and domestic transactions in PIIGS is made visible in figure 5.3. The three continuous lines show the number of domestic, foreign and intra-PIIGS acquisitions. It can be seen that the number of domestic transactions is more volatile than that of the foreign transactions. Domestic transactions decreased heavily during the financial; and euro-crisis while foreign transactions were less volatile, especially during the euro-crisis. The number of intra-PIIGS acquisitions remained small during the whole investigated period. The dotted line in figure 5.3 represents the percentage of foreign acquisitions on total acquisitions. In 2008 22.6% of total acquisitions in PIIGS had a foreign acquirer. This number increased until 25.9% in 2010, the start of the euro-crisis, to increase further to an ultimate high of 34.9% in 2013.
I expect that foreign acquisitions coming from regions which suffered less from the crisis increased relatively more during the euro-crisis. Therefore, foreign investors are divided into five regions (the regions are previously mentioned in chapter 4). Figure 5.4 shows the number of acquisitions in PIIGS with acquirers from each foreign region as percentage of total acquisitions (domestic, foreign and intra-PIIGS). Especially the percentage of transactions with a U.S. acquirer increased during the euro-crisis, from 5.4% in 2010 to 9.8% in 2013. Also acquisitions with an acquirer of “Rest of the World” increased during both the euro-crisis and the financial crisis. The percentage of intra-PIIGS acquisitions was lower during the financial- and euro-crisis than in the pre- crisis period. Intra-PIIGS investors experienced the same problems of the crisis as the domestic acquirers which can be a reason for the lower relative number of their acquisitions in other PIIGS countries.
0 40 80 120 160 200
Portugal Ireland Greece
18
It is much harder to see if the relative number of EU7 and “Rest of Europe” acquirers increased during the crisis compared with pre-crisis levels. These percentages fluctuate more heavily and show a less frequent pattern. It is needed to test if during the crisis significantly more acquisitions are completed by foreign firms. This will be done in the following sub-paragraphs.
Figure 5.3, Domestic, foreign and intra-PIIGS acquisitions in PIIGS
Figure 5.4, Foreign acquisitions per region, % of total acquisitions (domestic and foreign)
5.1.3 Financial vs. Non-financial
Besides the region of origin also the sector of the acquiring firm is of importance. As stated in the literature, the fire-sale hypothesis leads to the expectation that foreign financial firms take more
0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 30.00% 35.00% 0 250 500 750 1000 1250 1500 1750 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total Domestic Total intra-PIIGS
Total Foreign Foreign % of Total
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
19
advantage of the assets at fire-sale prices than foreign non-financial firms since it is expected that they want prefer to increase their returns and/or sell the acquisition back to domestic investors when the crisis is ended. Table 5.2 shows the number of foreign and domestic financial and non-financial acquirers in PIIGS. During the crises, domestic acquisitions had in equal times a financial acquirer as a non-financial acquirer in comparison with the pre-crisis period. In the contrary, during the crises foreign acquirers were far more often of the financial sector than of the non-financial sector if compared with the pre-crisis period.
Table 5.2, financial and non-financial acquirers in PIIGS11
Pre-Crisis Euro Crisis
Domestic and intra-PIIGS Freq. Perc. Freq. Perc.
0 Non-financial 7578 59.9% 2105 59.6%
1 Financial 5079 40.1% 1425 40.4%
Foreign Freq. Perc. Freq. Perc.
0 Non-financial 2985 70.3% 927 63.2%
1 Financial 1263 29.7% 539 36.8%
5.2 Regression Analyses – Per Region
5.2.1 Methodology
I follow the methodology used by Alquist (et. al., 2013). To test that the composition of the transactions of foreign investors changed during the crisis I use a linear probability model in which the dependent variable is a dummy variable ( . The dummy variable can take a value of 1 if the transaction belongs to the category “foreign acquirer” (or “foreign acquirer of a specific region”) and the value 0 otherwise. The model to test the first hypothesis is estimated as:
( | ) (5.1)
In this model, k, j, c and t stand respectively for the transaction, the sector of the target, country of the target and the year of the acquisition. The most important explanatory variable is which is a dummy variable for crisis years, thus has a value of 1 in the crisis period and 0 otherwise. The other explanatory variables are , the percentage stake of the target which is acquired and , country×target-industry
fixed effects. The country×target-industry fixed effects are necessary to identify the crisis effect since this variable captures the effect of foreign acquirers always more active in a certain country and sector. When
20
the crisis occurs in these countries the estimated coefficient on the country×target-industry level will ascribe part of the effect of the crisis (Alquist et. al., 2013). The last explanatory variables are the control variables, , which includes indicators for each control variable (lagged four quarters) and all its interactions. The included variables are per country annual-GDP, per country annual-GDP-growth, quarterly euro-depreciation and the use of IMF credit and loans as a percentage of a country’s quota. The last factor in equation 5.1, , is the error term. To check for multicollinearity between the control
variables and the crisis dummy I constructed a correlation matrix (Appendix C). This matrix shows there is no evidence that combinations of explanatory variables seem to correlate12, which minimizes multicollinearity issues.
Although logit and probit models are discussed to be better methods when a limited dependent variable is used, I used a linear probability model instead. In probit models the maximum likelihood is biased when fixed effects, in this thesis country×target-industry fixed effects, are taken into account which is also called the incidental parameter problem (Greene, 2004). A logit model could also be useful for this research but these models suffer sometimes of a large number of right hand side variables (Alquist et. al., 2013). Besides this, the ease of Linear Probability Model outstands the negative effects since all the predicted probabilities lie between 0 and 1.
5.2.2 Results
Table 5.3 shows the estimated coefficients of the linear probability model. Only the variable of interest, the crisis dummy, is reported since the coefficients of the control variables are less relevant for the fire-sale hypotheses. The crisis dummy variable coefficient estimates if the probability that an acquisition is a foreign acquisition significantly increased during the crisis compared to total acquisitions. I reported the coefficients for both the financial crisis (2008-2013) and the euro-crisis (2010-2013) separately. For the tables in this thesis, each “row” refers to the line containing the estimated coefficients for the crisis dummy ( .
The first row shows the crisis dummy coefficients for acquisitions with a foreign acquirer. The coefficient declares that during the euro-crisis (second column) the chance significantly increased that an acquisition had a foreign acquirer. The financial crisis coefficient (first column) is smaller and less significant than the euro-crisis coefficient. This means that during the euro-crisis the probability that an acquisition had a foreign acquirer increased more than during the financial crisis as a whole.
12
21
Table 5.3 Increase in inward acquisitions in PIIGS during the crisis per region
Acquirer Fin crisis
2008-2013 Euro-crisis 2010-2013 Foreign 0.025* 0.054*** (0.01) (0.01) No. obs. 21,901 21,901 R² 0.05313 0.054 Intra-PIIGS -0.007* 0.001 (0.00) (0.00) No. obs. 21,901 21,901 R² 0.038 0.038 EU7 0.014** (0,01) 0.000 (0.00) No. obs. 21,901 21,901 R² 0.009 0.009 Rest of Europe -0.007 0.009 (0.01) (0.01) No. obs. 21,901 21,901 R² 0.045 0.046 US 0.004 0.019*** (0.01) (0.01) No. obs. 21,901 21,901 R² 0.033 0.034 ROW 0.013*** 0.026*** (0.00) (0.00) No. obs. 21,901 21,901 R² 0.013 0.014
Notes: Coefficients marked *, ** and *** are significant at respectively the 10%, 5% and 1% level. The reported dummy variable coefficients are for the dummy indicating if the transaction is completed during the financial crisis or the euro-crisis. The second lines represent the standard error. The coefficients off the other variables are not included in this table to prevent an overload on information, these results are available upon request. “Foreign” excludes intra-PIGGS acquisitions.
The second row does the same but than for intra-PIIGS acquisitions (compared to total acquisitions), the third row for acquisitions of EU7 acquirers, and so on. The table shows us some interesting results. Overall the likelihood of being acquired by a foreign firm increased during both the financial crisis and the euro-crisis. Row 2, 3 and 4 tell that this is not significant for acquirers of other European countries (intra-PIIGS, ROE and EU7) during the euro-crisis and only slightly significant for acquirers of EU7 countries during the financial crisis. The other European countries suffered of the crisis as well and
13
22
therefore it is not a surprise that investors of these countries did not relatively increase the number of their investments during the crisis.
In the contrary, during both crisis periods the likelihood of being acquired by a ROW-acquirer increased significantly while the chance of being acquired by a US-firm increased significantly during the Euro-crisis but not during the financial Euro-crisis as a whole. These results are in line with the expectations that regions which suffered less of the crisis, rest-of-the-world and U.S., acquired relatively more in PIIGS14 during the euro-crisis.
5.3 Regression Analyses – Financial vs. Non-Financial Sector
5.3.1 Methodology
As mentioned before, the sector of the foreign acquirer is expected to be of importance for the fire-sale hypothesis. Acquisitions by foreign financial firms are expected to increase during the euro-crisis since these firms are expected to buy the assets at fire-sale prices for “fire-sale motives” as short term increase in returns or reselling the acquisition for a higher price after the crisis. For the ease of testing I separated the companies’ industries in two groups: financial and non-financial. To test, the same linear probability model as for the composition of transactions per region (equation 5.1) is used. The dependent dummy variable can now take, for example, a 1 if the transaction has a foreign acquirer of the financial sector and a 0 otherwise.
5.3.2 Results
The crisis dummy variable estimates in row 1 and 2 of table 5.4 show that both foreign financial and non-financial acquiring firms significantly acquired more in PIIGS during the euro-crisis15. The coefficient of foreign financial firms shows that the probability that an acquisition had an acquirer of a foreign financial sector increased by 1.9% percent during the euro-crisis. The probability that an acquisition had a foreign non-financial acquirer significantly increased by 3.4% during the crisis. Striking is that the probability that an acquisition had a foreign non-financial acquirer increased more during the crisis than the probability that the acquisition had a foreign financial acquirer.
For the financial crisis period no significant evidence is found that foreign financial firms acquired relatively more than before the crisis. The probability that an acquisition had a foreign non-financial acquirer significantly increased during the financial crisis. For both the financial and euro-crisis no
14 In Appendix D regression results are included for samples without Spanish targets and with only Spanish targets
to see the differences with the original sample. I checked for Spain since 44.7% of the targets are Spanish.
23
evidence is found that specifically foreign financial firms have taken advantage of the fire-sale prices and therefore, it is not expected that the foreign acquisitions should have a higher flipping rate after the crisis. Table 5.4, Increase in inward acquisitions in PIIGS during the crisis – financial vs. non-financial acquirers
Acquirer Target Fin crisis
2008-2013
Euro-crisis 2010-2013
Foreign Fin. WP All 0.008
(0.01)
0.019*** (0.01)
No. obs. 21,901 21,901
R² 0.061 0.062
Foreign Non-Fin WP All 0.017** 0.034***
(0.010) (0.01)
No. obs. 21,901 21,901
R² 0.082 0.082
Notes: Coefficients marked *, ** and *** are significant at respectively the 10%, 5% and 1% level. The reported dummy variable coefficients are for the dummy indicating if the transaction is completed during the financial crisis and the euro-crisis. The second lines represent the standard error. The coefficients off the other variables are not included in this table to prevent an overload on information. These are available upon request. “Foreign” excludes intra-PIGGS acquisitions.
5.4 Regression Analyses - Degree of Corporate Control Sought
5.4.1 Methodology
The fire-sale hypothesis states that during a crisis more controlling stakes are acquired by foreign firms since depressed asset prices make it more profitable to acquire control. An acquisition with a controlling stake is defined as an acquisition in which 50% or more of the stakes are acquired by the investor. Foreign financial firms are expected to take more advantage of the fire-sale prices than other foreign investors and therefore the increase in control sought of foreign financial firms is expected to increase relatively more. To test for an increase in the degree of corporate control sought of foreign acquirers in PIIGS during the crisis also a linear probability model can be used. The model to test for the second hypothesis is similar to the model for the first hypothesis although the explanatory variable of the percentage stake acquired is not a variable to control for anymore.
( | ) (5.2)
24
Table 5.5, Increase in control sought of acquirers in PIIGS during the crisis
Acquirer Target Financial crisis
2008-2013 Euro-crisis 2010-2013 All All 0.012 0.043*** (0.01) (0.01) No. obs. 21,901 21,901 R² 0.045 0.046 Foreign All 0.021 (0.02) 0.046** (0.02) No. obs. 5,714 5,714 R² 0.060 0.060
Foreign European All 0.002 0.034
(0.02) (0.03)
No. obs. 3,913 3,913
R² 0.068 0.068
Foreign Outside Europe All 0.057* 0.065*
(0.03) (0.03)
No. obs. 1,801 1,801
R² 0.073 0.073
Foreign Financial All 0.029 0.088**
(0.04) (0.04)
No. obs. 1,802 1,802
R² 0.074 0.077
Foreign non-financial All 0.017 0.038*
(0.02) (0.02)
No. obs. 3,912 3,912
R² 0.061 0.062
Notes: Coefficients marked *, ** and *** are significant at respectively the 10%, 5% and 1% level. The reported dummy variable coefficients are for the dummy indicating if the transaction is completed during the financial crisis and the euro-crisis. The second lines represent the standard error. The coefficients off the other variables are not included in this table to prevent an overload on information. These are available upon request. “Foreign” excludes intra-PIGGS acquisitions.
5.4.2 Results
25
Foreign acquirers of regions which suffered less of the crisis (U.S. and ROW) took advantage of the fire-sale prices to acquire a controlling stake while foreign acquirers of regions which suffered did not took this advantage.
Row 5 and 6 show the regression results divided by the sector (financial or non-financial) of the foreign acquirer. The probability that an acquisition with a foreign financial acquirer involves a controlling stake increases by 8.8% in the euro-crisis. This coefficient is higher and more significant than the coefficient of the foreign non-financial firm acquiring a controlling stake. It can be said that foreign financial acquirers take more advantage of the fire-sale prices to buy a controlling stake than non-financial acquirers. Additionally, table 5.5 shows that for the financial crisis period almost all coefficients are insignificant and therefore it cannot be said that acquirers (or a group of acquirers) relatively increased the amount of controlling stake acquired during this crisis.
6. Discussion
During the discussion I will relate the results to the literature and the stated hypotheses. I continue with the contribution of this paper to the theory. Then, I discuss the practical relevance and implications of this paper. Afterwards, I state the limitations of this thesis and I will mention possible areas for further research.
6.1 Theoretical Discussion
Table 6.1 combines the overview of the hypotheses and the additional literature with the results of this thesis on FDI data (chapter 4) and Acquisitions data (chapter 5). As in Acharya et al (2011), Alquist et al (2013) and Aguiar & Gopinath (2005), I find significant results for a relative increase in foreign acquisitions during the euro-crisis. The coefficient for the financial crisis period is smaller but still significant and positive. It can be stated that foreign investors increased its relative share in PIIGS (especially) during the euro-crisis. Therefore, hypothesis 1 can be accepted. During the euro-crisis in PIIGS, as in previous crises in other regions, foreign investors increased its share in total acquisitions in the crisis hit countries.
26
significantly increase its share in inward acquisitions. My results show that foreign financial investors did not take more advantage of the crisis than foreign non-financial investors, so hypothesis 2 is rejected. When looking at the increase of controlling stakes acquired during crisis by foreign firms, the previous literature found opposite evidence. Acharya et al (2011) found that during the Asian crisis foreign firms relatively bought more controlling stakes while the results of Alquist et al (2013) showed that foreign firms did not buy more controlling stakes during the Asian crisis and several other crises. My results are in line with the results of Acharya et al (2011) since I found that during the euro-crisis the probability increased that foreign investors acquired controlling stakes. This increase was higher and more significant for foreign financial acquirers than for foreign non-financial acquirers, which is in line with the hypothesis that foreign financial firms should take advantage of the crisis by buying more controlling stakes. As a result, I accept hypothesis 3. A reason for the contrast between my results and the results of Alquist et al (2013) on hypothesis 3 can be the difference between the ease of acquisition of a controlling stake before the investigated crisis. If it is already easy to acquire control before the crisis started, the probability that foreign acquirers increase the amount of controlling stake acquired should increase less during the crisis than when it was hard to acquire control before the crisis.
Table 6.1 Overview of the hypotheses, previous theoretical evidence and this paper’s results
Hypothesis Results Previous Literature
FDI Data Acquisitions Data Aguiar & Gopinath Acharya et al Alquist et al H1 Relative increase in acquisitions
with a foreign acquirer Yes Yes Yes Yes Yes
H2 Relative increase in acquisitions
with a foreign financial acquirer - No n.i. n.i. No
H3 Relative increase in control
sought - Yes n.i. Yes No
H4 Relative increase in acquisitions with a foreign acquirer of regions which did not suffer of the euro-crisis
Yes Yes n.i. n.i. n.i.
27
As last, I investigated the increase in the probability of acquisition during crisis periods per region of the foreign acquirers. I hypothesized that regions which did suffer less of the crisis increased its relative share in inward acquisitions in the crisis hit countries while the regions which did suffer as well did not increase its relative share. For the euro-crisis, I found that the probability that an acquisition had an acquirer of a region which did not suffer of the crisis, i.e. the U.S. or “rest of the world”, increased significantly. The probability that an acquisition had an acquirer of regions which also suffered of the crisis, i.e. EU7 and intra-PIIGS, did not change during the euro-crisis. This evidence is in line with the expectations and therefore hypothesis 4 can be accepted.
Taken together, evidence for fire-sale FDI is found. Foreign investments relatively increased during the crisis, especially coming from regions which did not suffer of the crisis. Since this increase is not particularly coming from foreign financial firms, it is not fully expected that the reasons for the increase in acquisitions are for fire-sale motives.
6.2 Contribution
This paper contributes to the literature, first by adding an extra hypothesis to the fire-sale FDI theory saying that acquirers from regions which do not suffer of the crisis relatively increase their acquisitions in PIIGS during the crisis. Secondly, I have tested the fire-sale hypothesis for the PIIGS during the recent financial- and euro-crisis. Previous literature on this theory was mainly built on the Asian crisis in 1997 and the Latin American crisis in 1995.
Calderon and Didier (2009) stated that fire-sale FDI should not have taken place during the financial crisis since all countries suffered of the crisis. However, I found evidence for the fire-sale FDI theory during the crisis. The probability that an acquisition had a foreign acquirer increased during the financial crisis, especially for ROW acquirers. The results were even stronger for the euro-crisis period. Desbordes and Wei (2012) argued that fire-sale FDI should be a concept for emerging; instead of developed economies. However, my results show that also developed economies, here the PIIGS, faced an increase in foreign investors during the financial and euro-crisis.
6.3 Practical Relevancy
28
To see if the increase in foreign acquisitions has positive or negative effects for the PIIGS economies and its firms, the reason for the increase in foreign acquisitions is important. I did not find evidence for a larger increase of foreign financial investments than of foreign non-financial investments. Since foreign non-financial investors are supposed to focus on long-run investment plans rather than on acquiring for fire-sale motives, it can be discussed that these non-financial investors did not take advantage of the weakened targets but of the weak exchange rate or changed regulation. Foreign investments made for long-run motives ask for different actions of policy makers than foreign investments made for short-term exploitation of the country’s weakened situation.
6.4 Limitations
It is important to keep in mind that there are several limitations to my research. To fully investigate the fire-sale FDI hypothesis, data is needed on the flipping rates of acquisitions made in pre-crisis and crisis periods. This data is not available yet because no after-crisis years have been passed to investigate the concept of “flipping the acquisition”. I can only set up predictions for this crisis’ flipping rates deriving of hypothesis 3 (about financial and non-financial acquirers). As a result of the lack of data on flipping rates, the conclusions in this paper are far of complete. However, this research is a first and big step in the investigation of fire-sale FDI in PIIGS during the euro-crisis.
Even though I combined data on the magnitude of inward FDI flows/stocks and data on the number of acquisitions, it remains hard to draw conclusions on the monetary value of the acquisitions. According to the fire-sale FDI theory, foreign investors buy domestic assets at fire-sale prices. Since data on the price paid for the acquisition is lacking for most acquisitions due to the fact that these prices are not disclosed, it is unknown if these assets were actually at fire-sale prices.
7. Conclusion
29
PIIGS acquisitions while firms of regions which suffered of the crisis (EU7 and other PIIGS) did not increase their share.
The fire-sale FDI literature mentions that the increase in foreign acquisitions during a crisis is the result of foreign investors taking advantage of the liquidity problems of the domestic firms. They take this advantage for short-term revenue increases or to resell the acquisition after the crisis for a higher price, which they called “flipping the acquisition”. Specifically foreign financial firms are expected to increase their activity for these fire-sale motives. Due to the fact that flipping-rates were not available yet during my research, I used data on the sector of the foreign acquirer to make predictions about foreign investors flipping their crisis-time acquisitions in PIIGS. A regression analysis showed that both foreign financial and non-financial firms increased their share in acquisitions in PIIGS during the euro-crisis. The coefficient was larger for non-financial; than for financial firms. No evidence is found that specifically foreign financial acquirers took advantage of the crisis; on the contrary it seems that non-financial acquirers took more advantage. Therefore, the increase in foreign acquisitions are predicted to be not specifically for fire-sale reasons.
As last, I focused on the controlling stakes acquired by foreign firms during the crisis. According to the fire-sale FDI theory I expected that foreign firms took advantage of the crisis by acquiring controlling stakes. The results found by the regression analyses are in line with this hypothesis. Here, especially foreign financial acquirers and acquirers of regions which suffered least of the crisis increased their relative amount of controlling stakes acquired.
Taken together, evidence for fire-sale FDI is found. Foreign investments relatively increased during the crisis, especially coming from regions which did not suffer of the crisis. Since this increase is not particularly coming from foreign financial firms, the reason for the acquisitions cannot be fully subscribed to fire-sale motives. As a consequence, I expect that flipping rates will not increase substantially after the crisis and that foreign acquirers thus remain owners of the acquired firms. I found that the amount of controlling stakes acquired by foreign firms increased during the euro-crisis, especially for financial acquirers and acquirers of regions which suffered least. This is in line with the fire-sale FDI theory.
7.1 Future Research
30
acquired the company to resell after the crisis for a higher price. Yet, data on the concept of “flipping the acquisition” is lacking since not enough time has passed after the acquisitions made during the crisis. Within a few years, scholars can investigate if acquisitions made during the euro-crisis are flipped more frequently in the years subsequent of the crisis.
Besides, in the recent future it can be researched if the increase in foreign acquisitions in PIIGS were crisis-time specific or remain large when the crisis ends. If after the crisis comes out that foreign investors acquire even more in PIIGS when the crisis is over, the concept of fire-sale FDI seems to be not relevant for this crisis. Thus, to complete the literature on fire-sale FDI in PIIGS during the euro-crisis I recommend scholars to extend this paper when after-crisis data on foreign acquisitions is available.
8. References
Acharya, V., Song Shun, H., Yorulmazer, T., 2011. Fire Sale FDI. Korean Economic Review 27:2, 163-202.
Alquist, R., Mukherjee, R., Tesar, L., 2013. Fire-sale FDI or business as usual? Working Paper, National Bureau of Economic Research.
Aguiar, M., Gopinath, G., 2005. Fire-sale foreign direct investment and liquidity crises. Review of Economics and Statistics 87:3, 439-452.
Baker, M., Foley, F., Wurgler, J., 2009. Multinationals as Arbitrageurs: The Effect of Stock Market Valuations on Foreign Direct Investment. The Review of Financial Studies 22:1, 337-369.
Bawden, T., Cooper, C., 2012 February 18. Everything Must Go! The Great European Fire Sale. The Independent. Retrieved from www.independent.co.uk.
Calderon, C., Didier, T., 2009. Will FDI be Resilient in this Crisis? Office of Regional Chief Economist, World Bank.
Desbordes, R., Wei, S.J., 2012. Foreign Direct Investment, Financial Development and the 2007-2010 Global Financial Crisis. Working Paper, University of Strathclyde.
Gibson, H., Palivos, T., Tavlas, G., 2014. The Crisis in the Euro Area: An Analytic Overview. Journal of Macroeconomics 39, 233-239.
31
Greene, W., 2004. Fixed Effects and Bias Due to the Incidental Parameters Problem in the Tobit Model. Econometrics Review 23:2, 125-147.
Krugman, P., 200. Fire-sale FDI. University of Chicago Press, 43-60.
Mundell R, 2012. The European fiscal reform and the plight of the euro. Global Finance Journal 23, 65-76.
Polita, V., Wickens, M., 2014. How the Euro Crisis Evolved and how to Avoid Another: EMU, Fiscal Policy and Credit Ratings. Journal of Macroeconomics 39, 364-374.
Poulsen, L., Hufbauer, G., 2011. Foreign Direct Investment in Times of Crisis. Working Paper, Peterson Institute for International Economics.
Schleifer, A., Vishny, R., 1992. Liquidation Values and Debt Capacity: A Market Equilibrium Approach. The Journal of Finance 47:4, 1343-1366
Shemer, N., 2012 April 4. Israeli firms compete in Greece fire sale. The Jerusalem Post. Retrieved from www.jpost.com.
32
Appendix A
Table A.1 All Dutch acquirers in PIIGS during the euro-crisis
Dutch firms invested in PIIGS during the euro-crisis
AEG Power Solutions BV J van Walraven BV
Aegon NV Jadeium BV
Ambiente Arancione Koninklijke DSM NV
ATS International BV Koninklijke Vopak NV
Ballingarry BV Makro
Brabant Alucast International Nivalis BV
CED BV Philips Electronics NV
CG International BV PPF Group NV
De Lage Landen International Reed Elsevier NV
Delta Lloyd Asset Mgmt NV Riwal BV
DIF Infrastructure III Royal Imtech NV
D-Marine Investments Stratos BV
EQT Infrastructure BV Tauw Group
Hunter Douglas NV TomTom NV
IMCD Group BV VConsyst BV
ING Real Estate Iberian Value VimpelCom Ltd
Invel RE (Netherlands) I BV WGA Sports BV
33
Appendix B
Table B.1 Annual domestic acquisitions and intra-PIIGS acquisitions per target country Domestic Acquisitions and intra-PIIGS acquisitions
Portugal Ireland Italy Greece Spain
year Number % Number % Number % Number % Number %
1999 91 7.5% 66 7.6% 299 5.5% 148 14.1% 439 5.7% 2000 138 11.3% 66 7.6% 408 7.5% 151 14.4% 454 5.9% 2001 109 9.0% 42 4.9% 344 6.3% 150 14.3% 449 5.9% 2002 97 8.0% 40 4.6% 242 4.5% 67 6.4% 444 5.8% 2003 54 4.4% 58 6.7% 329 6.1% 48 4.6% 589 7.7% 2004 58 4.8% 51 5.9% 321 5.9% 10 1.0% 352 4.6% 2005 83 6.8% 62 7.2% 440 8.1% 46 4.4% 364 4.8% 2006 76 6.2% 103 11.9% 473 8.7% 50 4.8% 603 7.9% 2007 125 10.3% 80 9.2% 463 8.5% 70 6.7% 655 8.6% 2008 131 10.8% 64 7.4% 485 8.9% 46 4.4% 946 12.4% 2009 61 5.0% 74 8.6% 454 8.4% 57 5.4% 532 7.0% 2010 81 6.7% 38 4.4% 367 6.8% 74 7.1% 554 7.3% 2011 43 3.5% 32 3.7% 329 6.1% 70 6.7% 516 6.8% 2012 39 3.2% 42 4.9% 263 4.8% 32 3.1% 405 5.3% 2013 31 2.5% 47 5.4% 206 3.8% 27 2.6% 334 4.4% Total 1217 100% 865 100% 5423 100% 1046 100% 7636 100%
Table B.2 Annual foreign acquisitions (excluding intra-PIIGS) per target country Foreign Acquisitions per target country
Portugal Ireland Italy Greece Spain
year Number % Number % Number % Number % Number %
34 Table B.3 Annual foreign acquisitions per acquirer region
Foreign Acquisitions per acquirer region
EU7 Rest of Europe U.S. Rest of the World
year Number % Number % Number % Number %
1999 169 8.3% 139 7.4% 105 9.0% 37 5.8% 2000 215 10.6% 166 8.8% 112 9.6% 25 3.9% 2001 145 7.1% 136 7.2% 48 4.1% 35 5.5% 2002 125 6.2% 80 4.3% 45 3.9% 28 4.4% 2003 97 4.8% 89 4.7% 56 4.8% 15 2.4% 2004 86 4.2% 103 5.5% 72 6.2% 19 3.0% 2005 120 5.9% 122 6.5% 72 6.2% 42 6.6% 2006 131 6.5% 141 7.5% 75 6.4% 40 6.3% 2007 192 9.5% 176 9.4% 86 7.4% 60 9.5% 2008 150 7.4% 175 9.3% 87 7.5% 75 11.8% 2009 151 7.4% 108 5.7% 62 5.3% 36 5.7% 2010 119 5.9% 139 7.4% 77 6.6% 56 8.8% 2011 116 5.7% 101 5.4% 89 7.6% 64 10.1% 2012 110 5.4% 116 6.2% 84 7.2% 50 7.9% 2013 105 5.2% 91 4.8% 97 8.3% 52 8.2% Total 2031 100% 1882 100% 1167 100% 634 100%
Table B.4 Annual domestic and foreign acquisitions per sector of the acquirer
Domestic and Intra-PIIGS Foreign
Financial Non-financial Financial Non-financial
year Number % Number % Number % Number %
35
Appendix C
Table C.1 Correlation coefficients of the independent variables. Correlation coefficients of the independent
