Printer Friendly

Insolvencies in Professional Football: A German Sonderweg?


There are many reasons to admire German football. (1) One reason, it is alleged, is that financial regulation effectively limits the incidence of financial distress among football clubs, in contrast to other countries. There are many studies on the financial stability of the German system in the scientific literature which reflect this perception (see Acero, Serrano & Dimitropoulos, 2017; Dietl & Franck, 2007; Frick & Prinz, 2006; Nicoliello & Zampatti, 2016; Rohde & Breuer, 2016). These studies tend to focus primarily on the top two divisions of the football pyramid, which are regulated by the German Football League (DFL). The "pyramid" is a term used to describe the hierarchical league structure characteristic of most national football competition--lower divisions (tiers) are connected to higher divisions via the system of promotion and relegation on sporting merit--typically with multiple regional leagues and more clubs in each tier at the lower levels (Eady, 1993). Only a small fraction of clubs (the "superstars") are to be found at the very top of the German football pyramid and 99% of German football clubs are in divisions below the top two tiers (36 out of 24,958 registered football clubs).

In this paper we identify 119 football club insolvencies across the top five tiers of the German pyramid that have occurred since 1994. The performance of the top clubs depends to some extent on the financial viability of lower tier clubs, because lower division clubs are incubators of talent in the national football system and every club contributes to talent identification in some way (Goke et al., 2014; Prinz & Weimar, 2018; Wallrafen et al., 2018). Moreover, due to the system of promotion and relegation, more than one third of the insolvencies we identify relate to clubs that had played in one or both of the top two divisions over the previous twenty years. The focus on the financial stability of clubs currently playing in the top two tiers of football appears to have created biased perceptions about the financial health of clubs in the German football pyramid as a whole.

The financial instability of European football is well known and there is a long history of football club insolvency. The problem of financial instability was first noted by Sloane (1971, p. 122), and Szymanski (2015) cites numerous instances of insolvencies in England in the 19th century and between 1920 and 1939. Insolvencies rose sharply in England during the 1980s (Beech et al., 2010; Szymanski, 2017), while Scelles et al. (2018) document a regular pattern of insolvency in France from the 1970s onward, despite the establishment of the National Direction for Management Control (DNCG) aiming to prevent insolvencies in 1990 (Dermit-Richard, Scelles and Morrow, 2017). Barajas and Rodriguez (2010) report the troubled financial history of Spanish clubs and their frequent insolvencies, while Baroncelli and Lago (2006) discuss the bankruptcies among Italian football clubs. The incidence of insolvency among clubs in European leagues is documented by the European governing body, UEFA, in its annual Club Licensing Benchmarking Reports (2009-2015).

To investigate the financial stability of the German football system, this paper uses data drawn from official insolvency proceedings. First, descriptive statistics of 119 insolvency declarations since 1994 among the top five leagues in Germany are compared to similar statistics from England and France. In a second step, we empirically test how derivations from ex ante expectations about demand (spectators) and performance (league ranking) influence the probability of an insolvency for clubs in the top four leagues since 1994 (n=2,626). We develop a simple model to estimate adverse shocks in German football (including relegation) and compare our results to the estimates for English clubs from Szymanski (2017) and for French clubs from Scelles et al. (2018). The only significant difference we find between the German experience and that of France and England relates to the second tier. Insolvency in the top tier is very rare in all three countries (zero cases in Germany, two in England and three in France over the period 1992-2014). Insolvency is very common in all countries from the third tier down. The contrast between the second tier in Germany, and the third where insolvencies are more common, have led some German fans to christen the third tier "the death league." This is an interesting wrinkle, but overall we find extensive evidence of frequent insolvency in the lower divisions of German professional football, at a rate comparable to English and French professional football.

The reason for this shared experience, we argue, is the instability inherent in the promotion and relegation system. Relegation in particular entails severe adverse economic consequences for a club, which thus creates an incentive for clubs to stretch their finances to the limit to avoid this outcome. While clubs seek to balance their books in expectation, adverse shocks (e.g., injuries to key players, drop-out/insolvency of local sponsors) can lead to financial distress. We also find that adverse shocks closely relate to relegation events. In Germany, the restructuring of the third league (2008) and the insolvency of the sponsor/broadcaster Kinowelt AG/Sportwelt GmbH (2001) also had a significant impact on a club's probability of remaining solvent.

A German Football Sonderweg?

The concept of the "Sonderweg," or "special path," was advanced in the postwar era, mainly by German historians. According to the argument, Germany did not develop the institutions of liberal democracy in the 19th century, unlike most other western European nations, and this, uniquely, led to the triumph of totalitarianism (Fischer & Fletcher, 1986). The main argument for considering a German "football Sonderweg" is that the football authorities developed institutions which uniquely, or exceptionally, enabled clubs to avoid the financial chaos common elsewhere in Europe. Just as the evidential basis of the Sonderweg has been widely challenged by historical scholarship (Blackbourn & Eley, 1984; Peukert, 1993), the evidence on insolvencies in German football challenges the received view.

There are many examples of the football Sonderweg argument to be found in the literature. Most seem to originate from German authors. For example, Frick and Prinz (2006, p. 64) claim that "the financial stability of the clubs is usually attributed to the licensing system practiced by the league's organization since the 1960s. In Germany, clubs are required to submit budgets for the forthcoming season, including forecasts of their expected revenues. This system ensures that there is continued control over costs, particularly wage costs." Likewise, Brand et al. (2013, p. 138) state that "the licensing procedures governing German football imposed relatively strict financial constraints upon clubs, mandating compliance with comparatively demanding regulations." Wilkesmann et al. (2011, p. 138) refer to "the strict licensing regime that prevents bankruptcy of clubs." Perhaps unsurprisingly, this view is endorsed by the DFL, the league authority, itself (DFL, 2015): "The Bundesliga's licensing process, self-imposed by the clubs as part of their league statutes, ensures that no club from the top two tiers would find themselves in a situation of being unable to complete a league campaign due to financial deficiencies."

This assessment has been widely accepted outside of Germany. For example, Storm and Nielsen (2012, p. 196), discussing UEFA's Financial Fair Play regulations, make the following observation:
UEFA's initiative is inspired by the recent developments in Germany and
France. Judged from the experiences in Germany where a strictly
enforced licensing system has been in place for several years, results
of tighter control are positive. Besides growing interest from
spectators and TV audiences, the clubs in the German Bundesliga are now
reducing their debt portfolios, increasing their revenues and some
clubs even making small surpluses.

Morrow (2013, p. 301) cites Germany as one leading example: "The UK does not operate as demanding a licensing system as found in several other European countries like France, the Netherlands or Germany."

Such arguments have led organizations like Supporters Direct, an influential pressure group in the UK, to advocate the German style regulation (Supporters Direct, 2011, p. 36):
Supporters Direct advocate a system akin to the regulatory licensing
regime employed by the Bundesliga, the professional football league in
Germany. This places severe penalties on clubs that are not financially
sustainable. If clubs fail viability tests, their professional licenses
are revoked and the club is relegated to the semi-professional leagues.
This deterrent has clearly worked, with no insolvencies in the
Bundesliga since its formation in 1963, in stark contrast to the record
in England.

All of these claims and arguments may not be strictly false, but they are misleading when one considers the football pyramid more broadly.

Insolvency Regulations in German Football

The German Football Pyramid

Until 1963, all football players in Germany were in principle amateurs, while leagues operated on a regional basis. In that year, the Bundesliga was founded as a national professional league, initially with 16 teams, increased to 18 in 1965. In 1974, the 2. Bundesliga was founded, initially with two divisions, north and south, each with 20 teams. In 1981, the two joined into a single national division of 20 teams and in 2008, a national third division (3. Bundesliga) was created. Below the Bundesliga, there are regional divisions whose clubs have the chance to enter the national level through promotion on sporting merit. These have been restructured a number of times, notably in 1994 and to a lesser extent in 2000, 2008, and 2012. The main effect of the restructurings has been to reduce the number of teams operating at the second, third and fourth tiers of German football, as shown in Table 1.

Although the total number of teams operating in the four top tiers rose between 1985 and 1992 from 582 to 700, the number has since fallen to currently 146. As the highest level of football has become more "national" in character, entailing greater travel related costs, the governing body has endeavored to reduce the number of clubs which might struggle to meet this financial pressure.

Historically, the governance of the entire system was in the hands of the German Football Association (DFB). However, in 2000 the top two divisions created the German Football League (DFL e. V.) with considerable autonomy, while retaining membership of the DFB (Wilkesmann et al., 2011). This reorganization was reminiscent of the English Premier League's secession from the English Football League in 1992. The DFL now took control of the club licensing system for the top two divisions, which had previously operated under the auspices of the DFB. Currently, the DFL statues (Satzung and Ligastatut) codify the licensing system for the top two divisions. The third professional division (3. Liga) is licensed by the DFB while teams in the semi-professional fourth and fifth leagues (Regionalliga and Oberliga) are under the supervision of the regional football associations (Landesverbande).

The German Professional Football System and Insolvency Regulations

All German football organizations (irrespective of the precise legal structure, such as eingetragener Verein/ registered association, GmbH/Ltd. or AG/PLC) are subject to the general insolvency law (InsO, see appendix for a description of the German Insolvency law). However, German football regulations control the consequences of an insolvency with regard to the playing rights of the clubs. Before the season 2015-2016, the opening of an insolvency proceeding automatically led to relegation to the next lower league in the following season. In 2015, the DFB changed its rules with regard to insolvency proceedings: once an insolvency proceeding has been opened, the club is punished by a nine point penalty in the relevant season of league competition. This does not mean automatic relegation, although the loss of points could lead to relegation ([section]6 No. 6 DFB Spielordnung; [section]11 No. 5 DFL Ligastatut). The DFL already changed this rule for the 1. Bundesliga and 2. Bundesliga in 2007. As before, a club that is liquidated is eliminated from the DFB football club register.

However, a team (player, staff, coaches, youth teams) is permitted to re-enter the football system if the club is re-founded (subsequently called the "successor club"). A large fraction of liquidated clubs have been re-founded as a new club with a similar name. For example, VfB Leipzig was re-founded as 1.FC Lokomotive Leipzig, 1.FC Amberg was re-founded as FC Amberg, and SV Weingarten was re-founded as SV Weingarten 2007. These club "copies" were permitted to resume competition in the lowest division of the pyramid, often incorporating former players, staff, and even board members.

In other cases, the successor club merged with a club in a higher division, in order to avoid starting from the bottom of the pyramid. In some cases, insolvent clubs were allowed (according to the rules of the regional football association) to create a successor club to compete just one division lower, rather than being forced to re-enter in the lowest division (e.g. Torgelower SV Greif/ Torgelower FC Greif).

A club that declares insolvency but is able to annul the declaration before the official insolvency proceeding is opened (because it is able to meet its debts) faces no penalties. We found that 16% of all declarations since 1996 were annulled after a few months, indicating that these clubs raised the necessary capital to end their liquidity crisis. The process following the initial declaration of insolvency is summarized in the flow chart (Figure 1).

The Causes of Insolvency in Professional Football

Since 2009, the governing body of European football (UEFA) has operated a financial regulatory system, known as Financial Fair Play, for clubs playing in the competitions that it organizes (primarily the UEFA Champions League and the UEFA Europa League). One of the stated aims of this regulation is "to introduce more discipline and rationality in club football finances" (UEFA, 2012, p. 2).

Bailouts and refinancing of insolvent clubs have been commonplace throughout Europe, and some economists have argued (Andreff, 2007, 2015; Storm and Nielsen, 2012, 2015; Franck, 2014, 2015) that this situation is reminiscent of the soft budget constraint (SBC) theory of Kornai (1979, 1986). They argue that clubs are willing to spend beyond their means, because they know they will be bailed out, either by the government (local or national), by the fans, (2) or by a wealthy benefactor (Terrien, Scelles & Durand, 2016). An important difference, of course, is that the Soviet enterprises, which Kornai (1979) analyzed, produced little if any value, and were not much liked; by contrast the local football club is a deeply beloved part of the local communities.

According to the SBC argument, clubs spend as much money as they can and take up as much credit as they can with a view to winning and, in the event of failure, they collapse. While such behaviour is often attributed to irrationality, the SBC argument in reality reflects a rational response to incentives. Critics of the insolvency crises also suggest that the "winner-take-all" nature of sporting competition (there can be only one champion) exacerbates the problem. Such competitive processes potentially lead to a kind of rat race (Akerlof, 1976; Frank et al., 1996), where everyone overinvests.

Szymanski (2017) proposes an alternative theory. Football clubs operate in an intensely competitive environment. There are large numbers of clubs that compete in the market for playing talent, which has become global. Ability is, to a significant degree, observable because players perform regularly in a public environment. Thus, wages tend to reflect marginal revenue products, club performance (in terms of league position) tends to reflect player expenditures and economic profits are driven to zero. Szymanski (2015) provides detailed evidence in support of these claims based primarily on financial accounts from English football.

The promotion and relegation system adds a significant element of uncertainty to this environment. Relegation from one division to another entails a substantial loss of support and therefore revenue. Not only do clubs lose attendance, they are also likely to lose other sources of revenue, such as TV rights and sponsorships (Schreyer et al., 2018). Hence, it is rational to spend to one's financial limit to avoid relegation, but in a stochastic environment, adverse shocks will cause some clubs to drop to a lower tier anyway. Adverse shocks could affect both the production side (e.g. players get injured) and the demand side (e.g. stadium attendance is reduced because of an economic downturn, or broadcast and sponsoring contracts fail to be renewed). (3) In this context, clubs may be pushed beyond their financial limit and thus into insolvency. This theory predicts that insolvency should be more likely when clubs experience negative shocks, when their performance is declining, and when teams are relegated. We are able to use our data to examine whether it is consistent with the theory. Specifically, we can model the relationship between team performance and attendance and test whether insolvency is more likely in the face of negative shocks and/or relegation. (4)

Empirical Analysis

Data and Descriptive Statistics

A consistent difficulty encountered in analyzing German football is the absence of detailed financial data at the club level (in contrast to many other countries such as England, France, Italy, Spain; see Peeters and Szymanski (2014)). However, we do know that, in common with other major European leagues, revenues of the Bundesliga have grown rapidly from [euro]1.09 billion in the season 2003-2004 to [euro]3.24 billion in the season 2015-2016, a compound average annual growth rate of 9.7% (DFL, 2017). This rapid growth is broadly in line with the experience of the other major European football leagues (Deloitte, 2018). Financial distress has therefore not been associated with general economic weakness, but rather with strength.

For our principal variable of interest, we gathered information on declared insolvencies of German football clubs, while playing in one of the top five leagues since 1995. To gather the data, we used press releases and online searches. (5) The data is summarized in Table 2.

Over this 23-year period, we found 119 declarations of insolvency just over five per season, involving 92 different clubs (among the multiple offenders, three clubs declared insolvency three times). In 20 cases the insolvency process was halted since the club was able to meet its outstanding liabilities before the court opened the insolvency procedure, leaving 99 cases where the court opened the insolvency procedure. We know of 57 cases where an insolvency plan was executed, writing off some debt and restoring the club to solvency. In 36 cases the club was liquidated, while 25 clubs of these cases "survived" as successor clubs. Between 1992 and 2014, the number of insolvencies observed in Germany in the top three divisions was thirty, compared to forty and thirty-nine in France and England respectively over the same period for the analogous divisions (Szymanski, 2017; Scelles et al., 2018). While the incidence of insolvency in Germany is somewhat lower, it is, nonetheless, a frequent event in German football. Figure 2 illustrates both the absolute and relative frequency of these events among the top five tiers. During our sample period, roughly 1% of clubs per year entered insolvency. While insolvencies occur in every year of our data, there was a wave of insolvencies around 2001-2003. This can be attributed to the financial failure of Kinowelt AG/Sportwelt Gmbh, a company that had multiple sponsorships with German football clubs. Its insolvency in 2001 caused both a liquidity crisis and financial failure for numerous clubs. There is also some evidence of an increase in insolvencies after 2008, when the league system was reorganized. In 2008, the 3. Liga was created as a national league and the number of clubs participating in the top four tiers was reduced significantly from 216 to 110. To facilitate this reorganization, an exceptionally large number of clubs had to be relegated, which was likely to create financial distress.

Table 2 also reports the tier that the club was playing in when declared insolvent and also the highest tier that it had ever played in. No club has ever entered insolvency while playing in the top division, and only two clubs have entered insolvency since 1995 while in the second division. In addition, we have identified 21 cases where teams that once played in one or both of the top two tiers have become insolvent after relegation to a lower tier. The list of clubs with a declaration of insolvency includes some of the great names in the history of German football, such as Alemannia Aachen, VfiB Leipzig, KFC Uerdingen, Rot-WeiB Oberhausen, Rot-Weiss Essen, FC Rot-Weiss Erfurt or SV Waldhof Mannheim.

We also collected data on attendance at league matches. (6) While fairly comprehensive attendance data is available for the top two tiers, we were only able to gather data on the third tier going back to 1988, and for the fourth tier only (partially) as far back as 1995. The first panel of Figure 3 shows the aggregated seasonal attendance for each tier, and the second panel shows the annual average attendance per club for each league. We take annual (seasonal) attendance to be the best indicator of demand for the clubs, since annual attendance closely correlates with annual ticket revenue.

As in most countries, there are large differences between the tiers. In 2016, the ratio of average seasonal attendance per club between the four tiers were 35:15:6:1. It is easy to imagine that relegation by a single tier would be likely to reduce revenues significantly. It is not unknown for clubs to be relegated in successive seasons (i.e., to drop by two tiers in a little over a year). Figure 3 illustrates the rising trend of attendance of recent years across all levels, though without capturing the effects of rising ticket prices, broadcast revenues, sponsorship and merchandising, of course. Average attendance per club has risen sharply in the bottom two tiers, while the number of participating clubs has fallen, causing aggregate attendance to appear stable.

The third main variable is league rank, which we use to proxy team performance over time. We should expect that a team playing in a lower division has a lower rank. Hence it seems natural that the top division teams are ranked by their league position (1-18), while, in the second tier, the first league position is given a rank of 19, the second 20, and so on to rank 36. Given that the lower tiers compete in parallel regional leagues, we denoted the rank of teams in two possible ways. The first approach extended the same method to lower tiers, so that when there were two regional divisions in the third tier, the team in the first league position of each division was given a rank of 37, the second a rank of 38, and so on. We call this the "unadjusted" method. However, this gives rise to the following anomaly: When Germany reformed the pyramid to create a third national league, a team that came, say, fifth in that league would get the same rank (41) as both of the teams that came fifth in the old third regional leagues, although arguably that was a lesser sporting achievement (i.e., coming fifth in one of two regional leagues is not as great an achievement as coming fifth in a single national league). Additionally, a team at the top of the fourth tier when there were regional leagues higher up the pyramid would be given the same rank as a team in the fourth tier when there were no regional leagues above it, even though in the latter case this amounts to a smaller pool of superior teams. Thus we calculated an adjusted league rank, such that for each league position the rank would reflect the number of teams above it in the pyramid. For instance, in the period when the third tier was organized as two regional leagues, we give first position in each league a rank of 37, second position in each league a rank of 39 and so on. We call this the "adjusted" method. This does impose the assumption that each regional league is of roughly equal quality, but we have no empirical basis to arrive at a more nuanced weighting system.

Descriptive Comparison with England and France

Table 3 summarizes the levels of insolvency in German, English, and French football. When looking at the incidence of insolvencies of the football pyramid from 1992-2002, the frequency in Germany (22) was very similar to England (19) and France (24). Since 2003, the number of insolvencies in the top three tiers dropped sharply in Germany (8), while there was a smaller drop in France (16) and an increase in England (25). The difference was principally accounted for by the persistent insolvencies in the second tier, notably in England, but also in France, and their complete absence at this level in Germany during this period. The number of insolvencies for third division clubs in Germany (8), does not differ as much from France (12) or England (13). While insolvency declarations decreased for the top three German leagues, the incidence in the fourth and fifth tier German leagues has increased since 2003. Overall, we identified 51 declarations of insolvency for Germany since 2003 and 53 for England. In other words, clubs in the German pyramid appeared on average no more immune from insolvency than English clubs, although the level of the pyramid at which insolvencies occurred was somewhat different. We are not able to compare the entire pyramid in France during this period due to lack of data on the French fourth and fifth tiers. While it appears that the introduction of the DFL might have led to more financial stability among first and, especially, second division clubs, it also appears that lower tier clubs suffered more under these changes. On balance, these two effects offset each other. In sum, the numbers depicted by Table 3 do not show convincing evidence for a German "Sonderweg," when looking at the top five divisions of the pyramid.

We now show that insolvency is associated with a fall in league rank on a scale that is likely to lead to relegation. Figure 4 shows the average league rank of clubs in the ten years leading up to insolvency and 7 years after. The solid line in Figure 4 represents rank measured on the unadjusted method and the dashed line shows it on the adjusted method. Both representations tell a very clear story: Over the ten years prior to insolvency, the clubs experienced a decline in league rank, which is relatively modest and steady up to three years before insolvency, when it becomes precipitous. The decline in the last three seasons, and the associated fall in revenues, creates financial pressures which are likely to stretch liquidity and threaten insolvency.

This narrative is also borne out by the attendance data. Teams lose support as performance deteriorates, and this effect is particularly marked after relegation. Figure 4 also shows the path of attendance leading up to insolvency. Ten years prior to insolvency, attendance averages around 4,000 and this appears relatively stable up to 5 years before insolvency (although this in itself may represent underperformance, given the rising trend of attendance illustrated in Figure 3). For around five seasons, prior to insolvency, attendance collapses and has almost halved by the insolvency date. Clearly, these clubs are likely to have experienced financial distress because of falling revenues. These patterns are very similar to those found in Szymanski (2017, Figure 4) and Scelles et al. (2018, Figures 2-5).

Estimation Strategy for Insolvency Probability and Negative Shocks

Although there are very similar descriptive patterns of financial distress between Germany and other European football markets, there might be differences in the causal mechanism. Szymanski (2017) and Scelles et al. (2018) examined the impact of negative shocks on the probability of an insolvency. To compare their results to the case of Germany, we gathered a data sample from the top four German divisions for the seasons 1995-1996 to 2016-2017 (attendance data for the fifth division was not available in early years of our observation period), including league membership and league position at the end of the season, average attendance for the season, whether the club was promoted or relegated in the previous season, and a dummy variable indicating if a club declared insolvency during the season. Following Szymanski (2017) and Scelles et al. (2018), our argument is that underperformance of a club relative to expectations is likely to be reflected in falling demand, which can be exacerbated further still by relegation. These effects are likely to lead to falling revenues, creating liquidity pressures, which increase the probability of insolvency.

Underperformance relative to expectations can be described as negative shocks, which we can model with our data. In our model, attendance is assumed to reflect demand, which is in turn dependent on league performance, measured by rank. We use regressions to estimate the relationship between demand and performance and define shocks as the deviations of actual demand from expected demand, i.e., the residuals from the demand equation. We model financial distress (declarations of insolvency) as a function of adverse shocks (negative residuals) to team performance on the field, which may accumulate over time (i.e., lagged as well as current negative residuals). In addition, we allow for the possibility that relegation itself may viewed as a negative shock and therefore contribute to likelihood of an insolvency declaration. We use the log odds of league rank to proxy performance. This transformation provides an easier interpretation and can be justified on the grounds that the transformation enhances the marginal difference between ranks as you rise up the leagues, reflecting fan preferences (the difference between 1st and 2nd is much greater than the difference between 10th and 11th). Since rank is by its nature a relative concept, we also model demand in relative terms. Thus demand (relative attendance) for club x in league i and season t is defined as the average number of spectators of club x in season i multiplied by the number of home games (which varies between 28 and 38) divided by the sum of all spectators across all four leagues in season i.

Our estimation procedure naturally involves two steps--estimating the demand relationship and then using the residuals from that estimation to model insolvency in the second stage. Our estimating equations are thus defined as (1) and (2) below.

Demand Estimation: Relative [] = [alpha] + [[beta].sub.1]Relative [Attendance.sub.i,t-1] + [[beta].sub.2]LN Odds [Rank.sub.i,t-1] + [[beta].sub.3-5][Division.sub.i,t] + [[beta].sub.6] [Promotion.sub.i,t-1] + [[beta].sub.7][Relegation.sub.i,t-1] + [eta] (1)

Insolvency Estimation: Insolvency = [alpha] + [[beta].sub.1] Residuals Demand Estimation + [[beta].sub.2] [Promotion.sub.t-1] + [[beta].sub.3] [Relegation.sub.t-1] + [[beta].sub.4-5]Division + [[beta].sub.6-11] Region + (2)

[[beta].sub.12]Kirchmedia + [[beta].sub.13]Introduction 3rd. Divison + [eta]

The demand equation (Eq. 1) includes, alongside league rank, dummies for promotion in t-1 and relegation in t-1 and the division in which the club plays at time t. To allow for unobserved heterogeneity associated with the clubs, we used linear regression approach with club fixed effects (supported by Breusch-Pagan and Hausman test statistics). We report our results using the stepwise inclusion method in Table 4 and thus present four reduced (1-4) and one full model (5). Standard errors are clustered at club level to control for club dependent variance.

Following the first stage regressions, in the second stage regressions (Eq. 2) the bivariate insolvency variable is regressed on the residuals from the first stage model. Since the information on insolvency is bivariate, we estimated average marginal effects of a probit regression (Wilson & Wright, 2013). The data sample between the first stage and the second stage is reduced, since no variance in insolvency probability has been recorded for the first division clubs, which were thus excluded. Apart from the residuals, we include in the second stage estimation dummies for two exogenous shocks: the effect of the Kinowelt AG/Sportwelt GmbH bankruptcy on sponsored clubs, and the 2008 league reform, each of which were known events that raised the probability of insolvency, as we argued above. We also control for relegations and promotions as explicit (negative and positive, respectively) shocks. Furthermore, division and regional dummies were included. To test for cumulative effects, we included variants which allowed the impact of residuals in the current year (t), over two years (t + t-1), three years (t + t-1 + t-2) and over four years (t + t-1 + t-2 + t-3). This is the same approach followed by Szymanski (2017) and Scelles et al. (2018). The results of the second stage regression are reported in Table 5.

Estimation Results and Discussion

Table 4 shows the coefficients and t-statistics from the demand regressions (1-5). Based on the lower AIC and BIC information criteria, the full fixed effect regression (5) showed the best fit with the variables. All five models accounted for a significant fraction of the within variance of the data ([R.sup.2]between 0.51 and 0.70). The lagged dependent variables are statistically significant at the 1% level with the expected sign, while the coefficient on lagged league rank is significant and of the expected sign only when the divisional dummies, which are highly correlated with rank, are excluded. The promotion and relegation variables are highly significant and of the expected sign.

To model football club insolvency, we used the residuals from the full fixed effects demand regression (5) as a proxy for economic shocks. We used the probit estimation method to model the probability that a club declares insolvency as a function of the residuals from the first step equation. In addition, we included dummy variables to allow for the impact of promotion and relegation, the division in which the club played, the region in which it was located (defined as regional associations within the DFB) and two exceptional events: the insolvency of the Sportwelt GmbH and the introduction of the third league in 2008. Columns (1) to (4) of Table 5 report the coefficients and z-values from the insolvency regressions. The coefficients reported are average marginal effects, which can be interpreted as partial elasticities. Column (1) includes only the residuals from the current period, column (2) the sum of current and lagged residuals, column (3) the summed residuals from t to t-2 and column (4) the summed residuals from t to t-3. Note that there is substantial attrition in the data as longer lags are added--over one third of observations are lost moving from column (1) to column (4).

We find a consistently negative relationship between the residuals and the probability of insolvency. This means that the more negative the shock to demand (residual), the higher the probability of an insolvency declaration. The coefficients are significant at the 1% level in columns (1) and (2), but not in column (3) and only marginally significant in column (4). We believe this is a result of data attrition. However, these findings are supported by the Figure 4, which suggests that the downward spiral toward insolvency is triggered a couple of years before the actual declaration. The pattern of results described here is in line with the results reported by Szymanski (2017) and Scelles et. al. (2018), who also find that negative residuals from the demand relationship increase the probability of insolvency.

We find that relegation in the previous season increases the probability of an insolvency declaration by 2.6 percent (based on column (1)); this result is statistically significant at the 1 percent level. We also find significant divisional effects. The reference group for the divisional dummies is the second tier. Even though we observe more insolvencies in the fourth division, our estimates suggest that, all else being equal, the hazard is actually higher in the third tier (the probability of declaring insolvency is 1.4% higher in fourth division relative to the second, but is 4.1% higher in third division compared to second division). This is explained by the fact that there are many more clubs at the fourth level than the third, and so while aggregate insolvencies are higher in number, the rate of insolvency is lower. This finding is also consistent with popular opinion among many German football fans, who refer to the third tier as the "death league." The gap between tiers, in terms of finances, varies, and it is plausible that the drop from the second to the third tier is particularly precipitous because of the extent of national exposure in the top two tiers. We find no evidence of regional differences in susceptibility to insolvency. In contrast, the two specific exogenous events are statistically significant. The Kinowelt and 2008 Reform coefficients are both positive and significant. The Sportwelt insolvency increased the risk of an insolvency by 5.6 percent and the reorganization of the third league by 5.8 percent (column (1) of Table 5). The latter case illustrates that club financial stability can be substantially affected by large scale restructuring of competition.

While we have uncovered many cases of insolvency in Germany over an extended period, our analysis has been constrained by the limitations of the available data. First, we have no financial information relating to revenues, wages, and profitability Were this available, we would be able to differentiate demand-side and supply-side shocks, as in Szymanski (2017). Second, the record of declarations of insolvencies may be incomplete, since no official insolvency data is available in aggregate form for the early years of our sample. Consequently, our results may underestimate the incidence of insolvency, especially in the lower leagues.


This paper challenges the widespread view that German football is characterized by financial stability and that German clubs are notably less susceptible to episodes of financial distress than clubs in other countries. We find that insolvency has been a persistent problem for many years and that the causes of insolvency are similar to those found in studies on football club insolvency in other countries.

We have identified 119 cases where football clubs among the top five tiers declared insolvency from 1995-1996 to 2017-2018, including 36 cases where a club was liquidated. To be sure, this pattern of insolvency has only affected clubs while playing at a level below the top two tiers. However, this is a pattern which is broadly comparable to that found for England (Szymanski, 2017) and France (Scelles et al., 2018), both in terms of the problem's scale and incidence. While there has been much talk of a German football "Sonderweg" (special, unique path) based on financial regulation and measures, such as the 50+1 rule, we find little evidence to believe that Germany is a unique case. True, insolvency is less common in the German second tier than in France or England, but in every other tier the patterns are similar. Insolvency is an extremely rare event in the top tier of these countries, and a very common event from the third tier down. In sum, clubs in the German football pyramid do not appear less susceptible to financial distress than those in England or France.

We have also found that the causes of insolvency can be traced back to the same phenomena found by Szymanski (2017) and Scelles et al. (2018): Adverse, unpredictable shocks in the relationship between team performance and attendance/revenues/resources and the event of relegation to a lower division. This in turn can be attributed to the severe consequences that follow from team underperformance and relegation. Underperforming clubs suffer from deteriorating league ranks, which quickly lead to relegation to a lower division and in turn provoke a severe liquidity crisis. This account differs from those we often see in theories that attribute failure to irrationality. Insolvency, in our view, is a consequence of institutional structure, not bad faith or incompetence. The finding that the league reorganization of 2008 substantially raised the probability of insolvency seems to underline this point.

While this study, together with those complementary studies of England and France mentioned above, suggests that there is a clear explanation of insolvency firmly rooted in the data and economic theory, there is no doubt to scope for further work. In particular, the lack of financial data for a large sample of clubs, which is available for English clubs, does suggest that future work would do well to extend the analysis in the German case if and when the relevant data becomes available.

Insolvency in football is an important issue since there has been strong support for financial regulation of football as exemplified by the Financial Fair Play regulations adopted by UEFA in 2009. While it is claimed that these regulations will impose "rationality and discipline," our research suggests that these regulations will have little effect, since they do not address the underlying source of the problem.

Our research has important policy implications. Some people believe that the high levels of insolvency among football clubs is a terrible problem which needs to be solved. Our research suggests that an obvious solution is to abolish or restrict the system of promotion and relegation, which is, we argue, the true source of financial instability. Indeed, some clubs have argued for such changes, but we believe that they more likely drew their motivation from a desire to enhance profitability than from the will to increase stability. We think that financial instability is a necessary corollary of a highly competitive market and that competition benefits the fans. Moreover, as we have pointed out, there is an added benefit in football, in that most clubs never really disappear if the business is liquidated--the fans just revive the club under a different name. During the period covered by this study, football has grown immensely in popularity in Germany, Europe and in the rest of the World. The competitiveness of the football market aided this growth, supported by the system of promotion and relegation (Pawlowski, Nalbantis & Coates, 2018; Scelles et al., 2016). From our perspective, financial instability is a feature, not a bug.

Appendix: The German Insolvency Proceeding

Until 1999, the German law of bankruptcy was called the Konkursordnung (KO). According to the KO, a firm that was unable to pay invoices due was declared insolvent. In consequence, the firm was liquidated and the remaining assets were sold, to pay back the creditors. The Konkursordnung was replaced by the Insolvenzordnung (InsO) in 1999, to save businesses that are commercially viable following a debt restructuring, with a view to preserving employment and maintaining output. Under the new law, the creditors of the firm can decide whether to approve an insolvency plan proceeding ([section][section] 217-269 InsO) or not, allowing the firm to survive after writing off or restructuring some of the debt.

In general, every organization (including non-profit community clubs (eingetragener Verein)) that is illiquid ([section] 17 InsO), facing a liquidity crisis ([section] 18 InsO, optional) or heavily indebted ([section] 19 InsO) has to inform the local court ([section] 13 InsO) about its solvency within a period of three weeks (Declaration of insolvency, [section] 15a InsO). The firm or an outside creditor can make the declaration of insolvency ([section][section]13 - 15 InsO). If the firm can fund the payment of outstanding invoices before the court has formally declared insolvency, the proposer of the insolvency declaration can annul the declaration ([section]13 InsO).

If the declaration is not annulled and the remaining assets of the firm are worth less than the expected costs of the proceeding, then the insolvency proceeding will not be opened and the firm will be liquidated ([section]26 InsO). If there are sufficient assets, the insolvency proceeding is opened ([section][section]27 - 30 InsO) and an external insolvency manager is announced by the local court ([section]56 InsO). Since 2012, the board of the firm itself can also act as insolvency manager in the opening-phase ([section]270 InsO). In due course, the local court calls all creditors and the insolvency manager to a general meeting ([section]74 InsO) and the insolvency manager or the firm is instructed to draw up a plan to restructure the firm ([section]218 InsO). If the creditors accept the plan to restructure the firm (which often includes a debt write off), the firm can proceed to execute the plan ([section][section]244-253 InsO). The temporary insolvency manager remains in charge until the insolvency plan is executed ([section]268 InsO). After a successful insolvency plan proceeding, the firm survives with reduced debt. If the creditor meeting rejects the insolvency plan presented by the insolvency manager, the remaining assets of the firm are disposed to pay out creditors and the firm will be liquidated ([section]231 InsO).


(1) The organization of German football is widely respected across the world for a number of reasons. The success of both the men's and women's national teams has been extraordinary, with six World Cup titles between them (four from 18 attempts for the men, two from seven for the women). This success is underpinned by a system of mass participation--in 2006, FIFA ranked Germany fourth in the world in terms of absolute numbers of players and second in terms of participation rate. The top national league (1. Bundesliga) has recorded the highest average attendance of any football league in the world for several years, and in Bayern Munich the league boasts one of the most successful and popular clubs in the world. By the standards of the top leagues in Europe, ticket prices are relatively low, a fact attributable in part to the "50+1" rule which requires that majority control of clubs to remain in the hands of association members who pay an annual subscription. Since the successful World Cup in 2006, Germany also has some of the best football stadiums in the world.

(2) One concrete example from Germany is the success of "fan bonds," which were sold out even when the clubs were in a precarious financial and sporting situation (Weimar & Fox, 2012).

(3) Feuillet, Scelles & Durand, 2018, consider cases where broadcasters have overbid for rights leading to financial failure--such cases can generate negative shocks for football clubs if a broadcaster defaults on a contractual payment.

(4) One referee suggested that the SBC and the "Szymanski" approach are parallel. This is a matter for debate, but here we wish to stress the difference. In the SBC approach the manager of the business deliberately breaches an officially imposed budget constraint, knowing that the constraint is not credible and will not be enforced. In our story, managers attempt to meet the budget constraint in expectation, but randomly fail. It is also worth pointing out that neither the SBC model nor ours involves "irrationality."

(5) Among our sources of information/news on insolvency proceedings, the most frequent pages with dedicated information were,,,,, club webpages, and regional newspapers, like or

(6) Attendance data was retrieved from,,, or (especially for lower leagues before 2005) via exclusive information submitted b e-mail from the administrators of, or exclusive information from regional football associations.


Acero, I., Serrano, R., & Dimitropoulos, P. (2017). Ownership structure and financial performance in European football. Corporate Governance: The International Journal of Business in Society, 17(3), 511-523.

Akerlof, G. (1976). The economics of caste and of the rat race and other woeful tales. The Quarterly Journal of Economics, 90(4), 599617

Andreff, W. (2007). French football: A financial crisis rooted in weak governance. Journal of Sports Economics, 8(6), 652-661.

Andreff, W. (2015). Governance of professional team sports clubs: Agency problem and soft budget constraint. In W. Andreff (Ed.), Disequilibrium sports economics: Competitive imbalance and budget constraints (pp. 175-227). Cheltenham, UK: Edward Elgar.

Barajas, A., & Rodriguez, P. (2010). Spanish football clubs' finances: Crisis and player salaries. International Journal of Sport Finance, 5(1), 52-66.

Baroncelli, A., & Lago, U. (2006). Italian football. Journal of Sports Economics, 7(1), 13-28.

Beech, J. (2014) Football Management Blog. Retrieved from

Beech, J., Horsman, S., & Magraw, J. (2010). Insolvency events among English football clubs. International Journal of Sports Marketing and Sponsorship, 11(3), 53-66.

Brand A., Niemann, N., & Spitaler, G. (2013). The two-track Europeanization of football: EU-level pressures, transnational dynamics and their repercussions within different national contexts. International Journal of Sport Policy and Politics, 5(1), 95-112

Blackbourn, D., & Eley, G. (1984). The peculiarities of German history: Bourgeois society and politics in nineteenth-century Germany: Bourgeois society and politics in nineteenth-century Germany. Oxford: Oxford University Press.

Deloitte (2018). Annual Review of Football Finance 2017. [website]

Dermit-Richard, N., Scelles, N., & Morrow, S. (2017). French DNCG management control versus European UEFA financial fair play: A divergent conception of financial regulation objectives. Soccer & Society, doi: 10.1080/14660970.20171323740

Dietl, H. M., & Franck, E. (2007). Governance failure and financial crisis in German football. Journal of Sports Economics, 8(6), 662669.

DFL (2015). The licencing process: How teams gain permission to play in the Bundesliga." [website] Retrieved from

DFL (2017). DFL-Report 2017 - Die wirtschaftliche Situation im Lizenzfussball. DFL Bundesliga [website]. Retrieved from from

Eady, J. (1993). Practical sports development. London, England: Pitman.

Feuillet, A., Scelles, N., & Durand, C. (2018). A winner's curse in the bidding process for broadcasting rights in football? The cases of the French and UK markets. Sport in Society, doi: 10.1080/17430437.2018.1505869.

Finch, V. (2009). Corporate Insolvency Law: Perspectives and Principles. 2nd Ed. Cambridge, England: Cambridge University Press.

Fischer, F., & Fletcher, R. (1986). From Kaiserreich to Third Reich: Elements of continuity in German history, 1871-1945. New York: Harper Collins Publishers Ltd.

Frank, R. H., Cook, P.J., & Rosen, S. (1996). The winner-take-all society. Journal of Economic Literature, 34(1), 133-134.

Franck, E. (2014). Financial fair play in European club football: What is it all about? International Journal of Sport Finance, 9(3), 193-217.

Franck, E. (2015). Regulation in leagues with clubs' soft budget constraints: The effect of the new UEFA Club licensing and financial fair play regulations on managerial incentives and suspense. In W. Andreff (Ed.), Disequilibrium sports economics: Competitive imbalance and budget constraints (pp. 228-249). Cheltenham, UK: Edward Elgar.

Frick, B., & Prinz, J. (2006). Crisis? What crisis? Football in Germany. Journal of Sports Economics, 7(1), 60-75.

Goke, S., Prinz, J., & Weimar, D. (2014). Diamonds are forever: Job-matching and career success of young workers. Jahrbucher fur Nationalokonomie und Statistik, 234(4), 450-473.

Kornai, J. (1979). Resource-constrained versus demand-constrained systems. Econometrica: Journal of the Econometric Society, 47(A), 801-819.

Kornai, K. (1986). The soft budget constraint. Kyklos, 39(1), 3-30.

Morrow, S. (2013). Football club financial reporting: Time for a new model? Sport, Business and Management: An International Journal, 3(4), 297-311.

Nicoliello, M., & Zampatti, D. (2016). Football clubs' profitability after the financial fair play regulation: evidence from Italy. Sport, Business and Management: An International Journal, 6(4), 460-475.

Pawlowski, T, Nalbantis, G., & Coates, D. (2018). Perceived game uncertainty, suspense and the demand for sport. Economic Inquiry, 56(1), 173-192.

Peeters, T, &Szymanski, S. (2014). Financial fair play in European football. Economic Policy, 29(78), 343-390.

Peukert, D. (1993). The Weimar Republic. London: Macmillan.

Prinz, J., & Weimar, D. (2018). The golden generation: The personnel economics of youth recruitment in european professional soccer. In: Neil Longley (Eds.), Personnel Economics in Sport. Edward Elgar.

Rohde, M., & Breuer, C. (2016). The financial impact of (foreign) private investors on team investments and profits in professional football: Empirical evidence from the premier league. Applied Economics and Finance, 3(2), 243-255.

Scelles, N, Durand, C, Bonnal, L., Goyeau, D., & Andreff, W. (2016). Do all sporting prizes have a significant positive impact on attendance in a European national football league? Competitive intensity in the French Ligue 1. Ekonomicheskaya Politika / Economic Policy, 11(3), 82-107. Retrieved from

Scelles, N, Szymanski, S., & Dermit-Richard, N. (2018). Insolvency in French soccer: The case of payment failure. Journal of Sports Economics, 19(5), 603-624.

Schreyer, D., Schmidt, L., & Torgler, B. (2018). Game outcome uncertainty and television audience demand: New evidence from German football. German Economic Review, 19(2), 140-161.

Sloane, P. J. (1971). The economics of professional football: The football club as a utility maximiser. Scottish Journal of Political Economy, 18(2), 121-146.

Storm, R. K, & Nielsen, K. (2012). Soft budget constraints in professional football. European Sport Management Quarterly, 12(2), 183-201.

Storm, R.K., & Nielsen, K. (2015). Soft budget constraints in European and US leagues: Similarities and differences. In W. Andreff (Ed.), Disequilibrium sports economics: Competitive imbalance and budget constraints (pp. 151-174). Cheltenham, UK: Edward Elgar.

Supporters Direct (2011). Developing football regulation to encourage supporter community ownership in football, Briefing Paper No 2. Retrieved from

Szymanski, S. (2015). Money and Soccer. New York: Nation Books.

Szymanski, S. (2017). Entry into exit: Insolvency in English professional football. The Scottish Journal of Political Economy, 64(A), 419-444.

Terrien, M., Scelles, N, & Durand, C. (2016). French 75% tax rate: An opportunity to optimize the attractiveness of the French soccer league. International Journal of Sport Finance, 11(3), 183-203.

UEFA (2012). Club licensing and financial fair play regulations. Edition 2012.

Wilkesmann, U, Blutner, D., & Muller, C. (2011). German football: Organizing for the European top. In Gammelsaeter, H. & Senaux, B. (Eds.). The organisation and governance of top football across Europe (pp. 138-153). London: Routledge.

Wallrafen, T, Pawlowski, T, & Deutscher, C. (2018). Substitution in Sports: The Case of Lower Division Football Attendance. Journal of Sports Economics, doi: 10.1177/1527002518762506.

Weimar, D., & Fox, A. (2012). Fananleihen als Finanzierungsmoglichkeit von Sportclubs? Eine Bestandsaufnahme am Beispiel der FuBballbundesliga, [Fan bonds as a financing option for sports clubs? An inventory based on the example of the football Bundesliga] Corporate Finance, 3(4), 181-187

Wilson, N, & Wright, M. (2013). Private equity, buy-outs and insolvency risk. Journal of Business Finance & Accounting, 40(7-8), 949-990.

Stefan Szymanski (1) and Daniel Weimar (2)

(1) University of Michigan

(2) University of Duisburg-Essen

Stefan Szymanski is the Stephen J. Galetti Professor of Sport Management at the University of Michigan. His main research interests are the economics of sporting contests and the financial analysis of football clubs.

Daniel Weimar is a postdoctoral researcher at Mercator School of Management (University of Duisburg--Essen). His research interests include sport economics, personnel economics, higher education economics, and media economics.
Table 1. Composition of the Top Tier German Football System Since

           1st Division  2nd Division  3rd Division    4th Division
Season     Name   Teams  Name   Teams  Name     Teams  Name  Teams

2016/2017  1. BL  18     2. BL  18     3. Liga   20    RL     90
2015/2016  1. BL  18     2. BL  18     3. Liga   20    RL     90
2014/2015  1. BL  18     2. BL  18     3. Liga   20    RL     90
2013/2014  1. BL  18     2. BL  18     3. Liga   20    RL     90
2012/2013  1. BL  18     2. BL  18     3. Liga   20    RL     90
2011/2012  1. BL  18     2. BL  18     3. Liga   20    RL     54
2010/2011  1. BL  18     2. BL  18     3. Liga   20    RL     54
2009/2010  1. BL  18     2. BL  18     3. Liga   20    RL     54
2008/2009  1. BL  18     2. BL  18     3. Liga   20    RL     54
2007/2008  1. BL  18     2. BL  18     RL        36    OL    144
2006/2007  1. BL  18     2. BL  18     RL        36    OL    144
2005/2006  1. BL  18     2. BL  18     RL        36    OL    144
2004/2005  1. BL  18     2. BL  18     RL        36    OL    144
2003/2004  1. BL  18     2. BL  18     RL        36    OL    180
2002/2003  1. BL  18     2. BL  18     RL        36    OL    180
2001/2002  1. BL  18     2. BL  18     RL        36    OL    180
2000/2001  1. BL  18     2. BL  18     RL        36    OL    180
1999/2000  1. BL  18     2. BL  18     RL        72    OL    160
1998/1999  1. BL  18     2. BL  18     RL        72    OL    160
1997/1998  1. BL  18     2. BL  18     RL        72    OL    160
1996/1997  1. BL  18     2. BL  18     RL        72    OL    160
1995/1996  1. BL  18     2. BL  18     RL        72    OL    160
1994/1995  1. BL  18     2. BL  18     RL        72    OL    160
1993/1994  1. BL  18     2. BL  20     OL       160    VL    496
1992/1993  1. BL  18     2. BL  24     OL       160    VL    496
1991/1992  1. BL  20     2. BL  24     OL       160    VL    496
1990/1991  1. BL  18     2. BL  20     OL       128    VL    496
1989/1990  1. BL  18     2. BL  20     OL       128    VL    416
1988/1989  1. BL  18     2. BL  20     OL       128    VL    416
1987/1988  1. BL  18     2. BL  20     OL       128    VL    416
1986/1987  1. BL  18     2. BL  20     OL       128    VL    416
1985/1986  1. BL  18     2. BL  20     OL       128    VL    416
1984/1985  1. BL  18     2. BL  20     OL       128    VL    416

Season     Teams

2016/2017  146
2015/2016  146
2014/2015  146
2013/2014  146
2012/2013  146
2011/2012  110
2010/2011  110
2009/2010  110
2008/2009  110
2007/2008  216
2006/2007  216
2005/2006  216
2004/2005  216
2003/2004  252
2002/2003  252
2001/2002  252
2000/2001  252
1999/2000  268
1998/1999  268
1997/1998  268
1996/1997  268
1995/1996  268
1994/1995  268
1993/1994  694
1992/1993  698
1991/1992  700
1990/1991  662
1989/1990  582
1988/1989  582
1987/1988  582
1986/1987  582
1985/1986  582
1984/1985  582

Note: 1. BL = 1. Bundesliga, 2.BL = 2. Bundesliga; RL = Regionalliga;
OL = Oberliga; VL = Verbandsliga. Season with restructure are

Table 2. Insolvency Statistics of German Top Tier Football Clubs
(1995-1996 to 2017-2018)

Division  Declaration  Annulled  Plan  Liquidation  Procedure still
                                                    open 2018

1           0           0         0     0           0
2           2           0         2     0           0
3          28           3        22     1           2
4          57          10        24    21           2
5          32           7         9    14           2
Sum       119          20        57    36           6

Tier  Tier that club played  Highest tier achieved before entering
      in at date of          insolvency (1994/95-2017/18)
1      0                      9
2      2                     35
3     28                     36
4     57                     37
5     32                      1

Table 3. Insolvency Declarations in European Football--A Comparison

                               1   2  3    4         5

1992-2003                      0   2  20   24        9
2003-2014                      0   0   8   29       14
England (Beech, 2014)
1992-2003                      0   8  11   19        6
2003-2014                      2  10  13   10       18
France (Scelles et al., 2018)
1992-2003                      2   6  16   no Data  no Data
2003-2014                      1   3  12   no Data  no Data

                               Sum (1-3)  Sum

1992-2003                      22         55
2003-2014                       8         51
England (Beech, 2014)
1992-2003                      19         44
2003-2014                      25         53
France (Scelles et al., 2018)
1992-2003                      24         24
2003-2014                      16         16

Table 4. Demand Fixed Effect Regression (1995-1996 to 2016-2017)

DV: Annual [attendance.sub.t]       (1)               (2)
                                    OLS FE            OLS FE

Annual [attendance.sub.t-1]          0.689             0.634
                                   (21.81) (***)     (18.90) (***)
Log Odds of [rank.sub.t-1]                             0.001
                                                     (5.92) (***)
Division (Ref=Third Division)
First League
Second League
Fourth League

Constant                            0.002              0.003
                                   (9.74)" (*)       (10.85) (***)
AIC                            -27781.209         -27829.567
BIC                            -27775.212         -27817.573
R2 between                          0.998              0.995
R2 within                           0.509              0.517
Observations                     2971               2971

DV: Annual [attendance.sub.t]        (3)              (4)
                                     OLS FE           OLS FE

Annual [attendance.sub.t-1]         0.458              0.680
                                  (15.10) (***)       (25.35) (***)
Log Odds of [rank.sub.t-1]         -0.001              0.001
                                  (-5.04) (***)       (6.87) (***)
Division (Ref=Third Division)
First League                        0.010
                                  (13.86) (***)
Second League                       0.003
                                  (10.43) (***)
Fourth League                      -0.001
                                  (-8.55) (***)
[Promotion.sub.t-1]                                    0.003
                                                      (9.85) (***)
[Relegation.sub.t-1]                                  -0.002
                                                     (-8.27) (***)
Constant                            0.003              0.003
                                  (12.27) (***)      (12.28) (***)
AIC                            -29042.329         -28684.743
BIC                            -29012.346         -28660.756
R2 between                          0.983              0.995
R2 within                           0.679              0.638
Observations                     2971               2971

DV: Annual [attendance.sub.t]       (5)
                                    OLS FE

Annual [attendance.sub.t-1]         0.522
                                  (18.11) (***)
Log Odds of [rank.sub.t-1]         -0.000
Division (Ref=Third Division)
First League                        0.007
                                  (10.24) (***)
Second League                       0.002
                                   (6.58) (***)
Fourth League                      -0.000
[Promotion.sub.t-1]                 0.002
                                   (8.51) (***)
[Relegation.sub.t-1]               -0.001
                                  (-5.00) (***)
Constant                            0.002
                                  (11.84) (***)
AIC                            -29242.271
BIC                            -29200.295
R2 between                          0.991
R2 within                           0.701
Observations                     2971

Note: t statistics in parentheses; Fixed effect regression with
standard errors clustered at club level, (***) p<0.01, (**) p<0.05,
(*) p<0.1

Table 5. Insolvency Probability Regressions

                                (1)               (2)
DV: Insolvency (0/1)
                                PROBIT-AME      PROBIT-AME

[Residuals.sub.t]                -12.525
                                 (-3.33) (***)
[Residuals.sub.t+t-1]                             -5.402
                                                 (-2.75) (***)
[Promotion.sub.t-1]               -0.021          -0.007
                                 (-1.45)         (-0.49)
[Relegation.sub.t-1]               0.026           0.023
                                  (2.83) (***)    (2.43) (**)
Division (Ref=Second Division)
 Third League                      0.041           0.038
                                  (4.62) (***)    (3.85) (***)
 Fourth League                     0.014           0.013
                                  (3.10) (***)    (1.94) (*)
Region (Ref=East)
 North                             0.008           0.015
                                  (0.76)          (1.24)
 South                            -0.003          -0.002
                                 (-0.47)         (-0.19)
 South-West                       -0.010          -0.007
                                 (-1.15)         (-0.76)
 West                              0.017           0.019
                                  (1.35)          (1.45)
 West-Berlin                      -0.009          -0.008
                                 (-0.84)         (-0.62)
 Sportwelt(2001)                   0.056           0.057
                                  (3.97) (***)    (3.70) (***)
 Introduction 3.Liga               0.058           0.059
                                  (4.28) (***)    (4.25) (***)
Pseudo-R2 Probit                   0.117           0.106
Observations                    2575            2200

DV: Insolvency (0/1)
                                  PROBIT-AME     PROBIT-AME

[Residuals.sub.t+t-1+t-2]          -2.152
[Residuals.sub.t+t-1+t-2+t-3]                       -2.144
                                                   (-1.68) (*)
[Promotion.sub.t-1]                -0.001           -0.000
                                  (-0.05)          (-0.02)
[Relegation.sub.t-1]                0.024            0.027
                                   (2.37) (**)      (2.50) (**)
Division (Ref=Second Division)
 Third League                       0.033           0.035
                                   (3.41) (***)    (3.34) (***)
 Fourth League                      0.018           0.017
                                   (2.20) (**)     (1.84) (*)
Region (Ref=East)
 North                              0.018           0.018
                                   (1.27)          (1.14)
 South                              0.001          -0.000
                                   (0.12)         (-0.00)
 South-West                       -0.004           -0.004
                                 (-0.38)          (-0.33)
 West                              0.022            0.019
                                  (1.42)           (1.11)
 West-Berlin                      -0.006           -0.008
                                 (-0.45)          (-0.55)
 Sportwelt(2001)                   0.054            0.056
                                  (3.29) (***)     (3.21) (***)
 Introduction 3.Liga               0.057            0.059
                                  (3.86) (***)     (3.66) (***)
Pseudo-R2 Probit                   0.093            0.102
Observations                    1896             1669

Note: t statistics in parentheses, standard errors clustered at club
level, (***) p<0.01, (**) p<0.05, (*) p<0.1 (*) p < 0.10,
(**) p < 0.05, (***) p<0.01
COPYRIGHT 2019 Fitness Information Technology Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Szymanski, Stefan; Weimar, Daniel
Publication:International Journal of Sport Finance
Article Type:Report
Geographic Code:4EUGE
Date:Feb 1, 2019
Previous Article:Jumping in the Pool: What Determines Which Players the NBA Considers in the Draft?
Next Article:Gender and Television Viewership of Professional Football.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters