Category: Blog

Is there a financial`fairness’ gap in English football ?

With the curtain closed on the 2014/15 Premier League season, the oligarch funded Chelsea reign supreme, looking over the shoulders at new money Manchester City.  This season also sees another Russian owner, Maxime Demin, financially facilitate Bournemouth’s emergence from the Championship wings to trip the boards on the main Premiership stage next season.  This emergence can also be attributed to Eddie Howth, Bournemouth’ young critically acclaimed manager, who has provided the vital strategic input for the playing success of the team. Bournemouth’s path to the top has be frought with financial instability, including entering into insolvency proceedings in 2008.  But will Bournemouth’s rise be fleeting and the struggle to stay up result in an overspend well beyond the income that they generate ?

Set against this backdrop is the continuing imposition of the Financial Fair Play (henceforth FFP) regulation by UEFA, whose controversial break-even rule is in reality an attempt to impose a measure of financial ‘fairness’ or efficiency on European Football clubs 1. Although recently it has emerged that there may be a softening of these rules, to date there have been 23 European clubs which have entered into settlements with UEFA as a result of the break even rule 2.

Myself and my colleague, Ronan Gallagher, are taking a professional interest in this financial regulation.  Academia (Sloane 1971; Késenne 1996; Garcia-del-Barrio & Stefan Szymanski 2009)  has long been aware that the business of football is best explained as maximising wins while just maintaining financial solvency although their financial (mis)management regularly makes headline news.  To this end we calculate one accounting measure of financial efficiency which is used to assess how effectively an institution can turn its spending into income.  Figure 1 displays the average wage to turnover ratio (%) along with 95% confidence intervals 3 for the Premiership and Championship clubs over the 2002/03-2013/14 period.  This one-dimensional measure of a club’s productivity has some major weakness4, but its exposition below illustrates some interesting features prevalent in the English game 5.

Figure 1


Firstly financial inefficiency, as measured by higher wage to turnover ratios,  has grown over the 2002/03-2012/13 period. This rising inefficiency is set against a backdrop of significant aggregate income growth of UEFA footballs teams of 45% 6 while European economies as a whole languished in a stagnant 1% growth period (Morrow ,2014).  Secondly, the 2013/14 season has seen a dramatic drop in financial efficiency especially in the Premiership.  This can largely be explained by two factors: 1) 2013/14 season was the first time club’s were subject to the Premier League’s Short Term Cost Control measure and also the first time that some club’s where subject to UEFA’s FFP break-even requirement; 2) 2013/2014 season seen a 29% year-on-year increase in total league revenue driven mostly (78% of the total increase) by the the first year of a new broadcasting rights package.7 .  Finally, there is a clear financial efficiency gap between the average Premiership and Championship club, with the latter having unsustainably high wage bills in the last few seasons (wage to turnover ratios of over 100%).  Furthermore, given the 95% confidence bands, this difference has become statistically significant from the 2007/08 season onward.8

In a recent paper, Goddard (2014) argues that this gap is likely due to the opening of competition through the promotion and relegation system.  He argues this system has a detrimental effect on profitability, owing to the pervasive tendency to overspend in an effort to achieve promotion or avoid relegation.  He points to the two tier system in English football as a point in case, where promoted teams fail to survive in the Premier league for more than one season, while regulated teams commonly experience financial duress upon arrival in the Football league.  Table 2 provides some evidence to suggest this is true.

Table 2 Financial efficiency implications for promotion and regulation for seasons 2002/03 to 2013/14.

 Mean Wage to Turnover Ratio (%)Standard ErrorLower 95% Confidence IntervalUpper 95% Confidence Interval
Promoted Clubs
Year Before78.03.670.585.4
Promotion Push Year 112.56.699.2125.8
Year After58.11.754.861.5
Relegated Clubs
Year Before89.35.278.799.8
Avoid Regulation Year66.13.359.372.9
Year After81.94.872.191.6

These results confirm that the club’s promotional push involves an unsustainable overspend, illustrated by an average wage to turnover ratio of 112.5% in the promotion push year. Furthermore the immediate revenue boost to these newly promoted clubs is evidenced by the dramatic and statistically significant drop in financial efficiency one year after their promotional campaign.  In contrast, relegated clubs experience a deterioration in financial efficiency the year they move into the Championship, indicated by an increase in the average wage to turnover ratio to 81.9% in that year.

While this preliminary analysis has some strong statistical health warnings we can attempt to provide some context to these findings.  The Premiership clubs are more efficient than Championship clubs because the TV money and associated sponsorship monies is a financial game changer.  Furthermore, the dramatic improvement in financial efficiency in 2013/14 is likely to be sustained with the dampening effect of cost control regulations extending beyond season 2015/16.  Therefore, clubs will bust a gut to get there and stay there.  Getting there takes a significant increase in financial input per unit output (i.e. a rise in your inefficiency in the year in which you go on the promotion push).  It’s a “go big or go home” scenario and many clubs fail to turn their spend into adequate points to ensure premiership survival.  Those that do “reach escape velocity” and stay in the premier league (e.g. the Stokes and Swanseas of the footballing world) ultimately achieve better long run efficiency but it’s high stake poker with more causalities than millionaires.

So Bournemouth’s (and indeed all the promoted teams) fight to survive (or perhaps thrive) in next season’s Premiership will provide for some interesting challenges for their future financial stability/efficiency.

Stata Code to create graph


Peeters, T., & Szymanski, S. 2014. Financial fair play in European football. Economic Policy, 29(78), 343–390. doi:10.1111/1468-0327.12031

Garcia-del-Barrio, Pedro, and Stefan Szymanski. 2009. “Goal! Profit Maximization Versus Win Maximization in Soccer.” Review of Industrial Organization 34 (1). Springer US: 45–68.

Goddard, John. 2014. “The Promotion and Relegation System.” In Handbook on the Economics of Professional Football, edited by John Goddard and Peter Sloane, 23–40. Edward Elgar Publishing.

Késenne, S. (1996). League management in professional team sports with win maximizing clubs. European Journal for Sport Management, 2(2), 14–22.

Morrow, Stephen, and Morrow Stephen. 2014. “Football Finances.” In Handbook on the Economics of Professional Football, edited by John Goddard and Peter Sloane, 80–99. Edward Elgar Publishing.

Sloane, Peter J. 1971. “THE ECONOMICS OF PROFESSIONAL FOOTBALL: THE FOOTBALL CLUB AS A UTILITY MAXIMISER.” Scottish Journal of Political Economy 18 (2): 121–46.


  1. See Syzmanski & Peeters (2014) paper for an excellent empirical exposition of this
  2.  Daniel Geey’s has a nice exposition of these settlements as well as a country breakdown
  3. based on a t distribution to account for small sample bias
  4. a) It assumes constant returns to scale. b) It doesn’t capture the non-profit nature of football clubs; for decades the economic study of football teams has long believed that they are run on a non-profit basis, a more appropriate model being win maximisation contingent on some budget constraint.  c) It suffers from the Fox’s paradox problem; a team could be partially efficiency in a number of areas but overall still performing poorly
  5. A notably, if obvious, feature that is illustrated by this simple ratio is highly labour intensive nature of the football business.  This is a feature of all sporting sectors
  6. According to Deloitte Annual Football Review data for the 2006-2010 period
  7. Information was taken from the  Deloitte Annual Review of Football Finance 2015
  8.  Generally when comparing to parameter estimates, such as sample means, it is always true that if their confidence intervals do not overlap then the statistics are significantly different. This simplistic statistical finding comes with some important caveats.  We assume that the mean are from independent sample and are approximately distributed normally.  The t distributions does help with some small sample bias but not with the independence assumption.

Financial risk taking and gender in my teaching

I have just finished reading The hour between dog and wolf by John Coates on the biology of financial risk taking. As a former derivatives trading Dr John Coates has first hand experience in the neurological patterns that traders experience during financial booms and busts and how this influences P&L account volatility.

The book is based on his years of academic research after returning to Cambridge University from industry to study a PhD in neuroscience and finance. He has since published considerable evidence refuting the neo-classical economic ideal that we choose our course of behaviour after thinking things through. 

 He argues that the aristotlean idea of the body and mind interconnectedness is a more coherent explanation for modern financial risk taking than traditional neck up economic rationalism.

The board tenant of the book is that the mind and body of traders works as a dynamic feedback loop, with testosterone being the molecule of irrational exuberance and a reason that ralllies change into bubbles.

One of Coates’ clear conclusion is that more female traders are needed to moderate the large fluctuations in financial markets.  This resonates with some antecdote evidence I have from teaching my Trading Principles course on the MSc in Computational Finance and Trading.  This course is based around real world trading simulations, and students are assessed on their trading book perfomance. 

This year a third of the students where female. This cohort performed much better in their trading book cumulative NAV (measured over all 16 simulations).  Also in the Law of One Price simuations the girls achieved much higher broker interest rebates, a clear indication that they are sucessfully offseting their risks from arbitrage trading.

Next year I hope to design a more scientific method to statistically confirm this antecdotal finding.


Why study a MSc in Computational Finance and Trading at Queen’s ?

I have created an online presentation (using prezi) with some adding voice over, explaining the key benefits to studying a MSc in Computational Finance and Trading in our new state of the art Riddel Hall campus at Queen’s Management School. This course has been designed to make you industry ready for a computational and or trading focused career, using teaching based on industry experience, academic excellence and cutting edge financial technologies.  Take a look at the below presentation for more details:


Financial data analysis as an iterative feedback process

In the first few lectures in my Financial Econometrics course I emphasise that analysing financial data should be an iterative learning feedback process, where an initial idea may be modified after some statistical analysis reveals an incomplete or incorrect answer to our initial hypothesis.  It is important for my students to understand that econometrics is not about producing ‘black box’ results from statistical software but is a gradual learning pathway which leads to asking the correct question using the correct empirical design and data.  This enables the analyst to avoid what Kennedy called Type III errors; getting a statistically precise answer to the wrong question.

Here is a short video explaining this process:

The Misery of the Technical Analyst

After reading Nate Silver’s excellent “The Signal and the Noise” book recently i thought i would illustrate the difficult that technical analyst (or chartists to be more derogatory) have in identifying the real information (signal) from the noise in UK stock markets.

I have taken the FTSE 100 as an example.  Below are six prices series.  Three of these are 1000 trading days of the FTSE100 in the 1990s, 2000s, and counting back from today.  The other three are fakes, and have been generated by simply flipping a coin (actually telling STATA to pick a random series of 1s and Os).

Technical analysis is identifying the signal solely on the basis of past statistical patterns, without consideration for the historical financial characteristics of a company.  You can have some sympathy for the difficult task they face in the graphs below, its really difficult to distinguishing the signal (FTSE100 series) from the noise (the fakes or random walks).



Click here to find out which graphs are the FTSE100 and how i used STATA to generate these graphs.


Is the Northern Ireland housing market in recovery ?

I have recently entered the N.Ireland housing market and have been taking a keen interest in its prospects.  Being a ‘number hugger’ I decided to look a little closer at the house sales data supplied by the UK National Statistics Office to assess if the market is really on the road to recovery after the 2007-2009 financial crisis and its unprecedented effect on the N.Ireland housing market.  I decided to attempt to predict where the market will go in the next few years using some commonly using statistical forecasting techniques.  As asset prices are notoriously difficult to predict I will look at the market depth (proxied by number of monthly transactions over £40K) as a indicator of the market’s recovery and build a statistically optimal forecasting model.

To assess this I use a monthly training data set over the period jan2008-dec2014.   I use the exponential smoothing modelling technique, a widely used forecasting technique in statistics. This technique is a weighted average of past observations that allows for forecasts to include the patterns in previous observations and can account for the additive or multiplicative nature of the seasonal, trend and noise components of a financial time series 1

The plots below shows the monthly total house sales in Northern Ireland over the Jan 2008- Sept 2013 period (red line), fitted values from the ETS models (blue line), and forecasts over the Jan 2015- Jan 2017 period (blue line).   The forecasts were produced using a Rob Hyndman’s excellent exponential smoothing package in R2.  Figure 1 shows a statistically optimal model, while Figure 2 shows a sub optimal model (where a multiplicative trend has been added) which has an upward trend in the forecasts. Both models have a multiplicative seasonal term and confidence intervals are calculated using a normal approximation.

Firstly both model show a strong seasonal trend in the market depth which is exactly what you would expect; its harder to sell your house in winter.   Interestingly, even though visually there seems to be a trend upwards in recent months, the optimising algorithm does not identify any trend term in the optimal model.  When I force the model to trend upwards in Figure 2 the forecasts are more visually appealing but the width of 95% confidence intervals suggest that these forecasts are vastly more inaccurate.  3

The optimal model (Figure 1) suggests a stabilised seasonal trend going forward but given the larger confidence intervals as we move into the future this forecast is far from accurate.

My analysis on forecasting NI house sales is far from exhaustive but I can draw some general conclusions.  Overall while the market has now stabilised in recent years, there are few signs of a sustained recovery in terms of the number of housing being sold above £40K.  It will be many years (if ever) before we will return to the irrational exuberance of house buying we seen in the lead up to the 2007-2008 crash.  Personally I hope such a market never returns.

FIGURE 1: Optimal Model using Hyndman et al 2002 algorithm


FIGURE 2: A suboptimal model with a multiplicative trend added


Click here for the R Code to recreate this plot.

  1. See Rob Hyndman’s excellent introductory chapter on the topic
  2. This package is based on Hyndman’s book on exponential smoothing and uses a statistically optimising algorithm to identify the appropriate model given the training dataset.
  3. For details of the modelling procedure see:

    Hyndman, R.J., Koehler, A.B., Snyder, R.D., and Grose, S. (2002) “A state space framework for automatic forecasting using exponential smoothing methods”, International J. Forecasting, 18(3), 439–454.

    Hyndman, R.J., Akram, Md., and Archibald, B. (2008) “The admissible parameter space for exponential smoothing models”. Annals of Statistical Mathematics, 60(2), 407–426.

Model Parameter Stability

The first assumption of the classic linear regression model is that the dependent (or outcome) variable can be calculated as a linear function of a specific set of independent (or predictor) variables, plus a disturbance term1.  This statement implies that the unknown parameters of this linear function are stable or constant over the estimation period.   This assumption is particularly important when using a linear regression model to predict.

In practical terms you are assuming  the effect of your predictor(s) remains unchanged over the sample period; for relationships between financial variables this may be unrealistic in the presence of large ‘landscape changing’ events.  In econometric terms these large events are sometimes called break points.

This instructive video explains in more detail how to test for parameter stability, both when a break point is known and when we cannot clearly identify a break point

Click here for the do file, and here for a copy of the slides.

  1. See Chapter 3 of Kennedy 1998 “A Guide to Econometrics” for an excellent introduction to the five assumptions underpinning classic linear regression models

My Belfast half marathon 2014

Last year I made my first tentative steps back into distance running.  I trained for the Belfast half using a technique called Chi Running, which I hope will allow me to run better and without injury into my 50s.  This technique focuses on your running form and takes a more holistic approach to running in general.  I am a long way from perfecting it but already i have noticed much fewer aches and pains during and after running.  I completely the Half marathon in a respectable time:

My longer term goal is to run one more full marathon (i have ran 3 big ones, London in 2000 (3.29.15), Dublin 2007 (3.15.01) and New York 2008 (3.39.15) ) under 3hrs 30mins using this new technique.

How does web technology adoption affect the performance of Irish credit unions?

Irish credit unions are now entering a period of substantive structural change which will in part be based around technological improvements. One of the conditions of the EU/IMF/ECB financial support package for Ireland was the requirement that credit unions are restructured. A Credit Union Re-Structuring Board was established in 2012 to facilitate amalgamations and the creation of strong ‘anchor’ credit unions capable of developing more sophisticated and more sustainable business models. €250 million has been allocated for this process some of which will be used to enable ‘anchor’ credit unions upgrade their ICT systems.  It was against this back drop, in our forthcoming paper[1], that we assessed the performance implications of web technology adoption in Irish credit unions over the 2002-2010 period.

Here are some key excerpts from the paper.

Figure 1 illustrates a steady increase in web adoption over the period although in 2010 53% of credit unions still did not have a web-based facility.  Further survey evidence from 2010/2011 suggests that these adopters are relatively unsophisticated technological capabilities, with less than 10% offering ATM and or phone banking. This could be related to two factors.  The first is that Irish credit unions have been unable to create a sophisticated integrated technology solution across credit unions and secondly credit unions are constrained by legislation and the regulatory authorities in the range of services that they provide

Figure 1


Figure 2 presents a preliminary visualisation of some performance and cost metrics grouped by whether the credit union has adopted a website or not. The graphical analysis reveals some distinct difference with web adopters experiencing lower spreads on average (driven it seems by lower average loans rates), a marginally higher pay-out ratio and higher average labour and capital expenditures. The latter finding is consistent with the initial encroachment on costs of the adoption of a new technology, while the former finding suggests that credit unions are passing any benefit accrued from this new technology to their membership. The graphical analysis suggests that both saving members and borrowing members benefit but that the majority of the benefit accrues to borrowers.

Figure 2


That said, care must be taken when drawing any causal inference on the effect of adoption from these graphs as to do so infers that the non-adopters and adopters have no other differences other than the adoption of a website.  Using an exhaustive econometric panel data analysis to account for both observable and unobservable difference between adopters and non-adopters, we find consistent statistical evidence of a reduction in spread ( driven by a fall in the loan rate) due to the adoption of web technology.  This effect is persistent over a two and three year period and translates into a cost benefit for borrowing members.

Overall our study highlights that the adoption of a website, even with limited functionality, can provide cost reductions and performance enhancement. This points to the potential of additional benefits accruing from more sophisticated levels of technological advance.  We feel this paper provides timely and important evidence to the Irish credit union sector, which is now entering a period of substantial structural change which is partly base around technological improvement.

1. McKillop, Donal G., and Barry Quinn. 2015. “Web Adoption in Irish Credit unions:Performance Implications.” Annals of Public and Cooperative Economics 86(3).