What do we know about factors determining debt structure of

companies?

Iwo Augustyński, PhD

Potential explanatory variables

• Firm fixed effects• Tax benefits• Size• Cash flow volatility• Assets tangibility/liquidity• Market-to-book ratio• Pecking order• Business cycle

Data and methodology

• Germany, France, Great Britain, Italy - four most advanced member countries, Greece and Spain – which suffer the most from crisis, Poland as the biggest economy from new member states and Finland as a representative of “northern economic model”. – 1618 firms in full sample

From the Amadeus dataset of Bureau Van Dijk I selected all firms meeting following criteria:• nonfinancial companies• independent

– or state-owned– or private or family owned– or foundation owned

• with data from whole period• indebted during all period

Variation in the leverage in the full sample

2004 2005 2006 2007 2008 2009 2010 2011

-10%

10%

30%

50%

70%

90%

110%

130%

mean+st. dev.-st. dev.Percentile 5%Percentile 95%minmax

LEV

Variation in the leverage in the group of big companies

2004 2005 2006 2007 2008 2009 2010 2011

-10%

10%

30%

50%

70%

90%

110%

130%

minmaxmean+st. dev.-st. dev.Percentile 5%Percentile 95%

LEV

Variation in the leverage in the group of small companies

2004 2005 2006 2007 2008 2009 2010 2011

-10%

10%

30%

50%

70%

90%

110%

130%

medianaminmaxmean+st. dev.-st. dev.Percentile 5%Percentile 95%

LEV

Average leverage ratio in the European countries

GB FI FR DE GR IT PL ES10%

15%

20%

25%

30%

35%

40%

45%

28%

20%

24%

27%

33% 33%

22%

31%33%

23% 22%

27%

40%

33%

27%

34%

28%

19%

22%

25%

45%

33%

27%

33%

2004 2005 2006 2007 2008 2009 2010 2011

Parameters of fixed effects panel models

  GB FI FR DE GR IT PL ES

Const. 0,70   -1,64   0,46 0,21 0,32  

Size     0,05     0,03    

Asset liq       0,43     0,28  

IR 0,53 0,88       0,26    

Tax benef     4,83   -0,63 0,22   0,91

Profit -0,11 -0,09 -0,04 -0,06        

GDP exp -0,40             -0,94

Debt str 0,03   -0,03   -0,05      

Adj R2 0,830 0,759 0,737 0,728 0,717 0,767 0,575 0,785

Adj R2 const* 0,800 0,714 0,709 0,698 0,655 0,762 0,449 0,771

Difference 0,030 0,045 0,028 0,030 0,062 0,005 0,126 0,014

* Models with only constant (β0) as a independent variableTable consist only variables with p-value over 0,05 and 0,01

Debt to fixed assets ratio of firms in selected European countries

GB FI FR DE IT PL10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

29.0%27.3%

31.6%

20.3% 20.8%

10.7%

36.0%34.6% 34.0%

19.5%

24.2%

17.1%

30.8%

34.1% 34.5%

18.3%

22.8%

19.5%

2004 2005 2006 2007 2008 2009 2010 2011

Debt to fixed assets ratio of firms in selected European countries. 2000-2011

Belgium

Czech Rep

ublic

Denmark

German

y

Estonia

Irelan

dFra

nce Italy

Cypru

s

Lithuan

ia

Hungary

Netherl

ands

Austria

Poland

Slove

nia

Finlan

d

United Kingd

om

Total

mea

n

Total

mea

n with

out IR an

d CY0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

25.2%

7.6%

21.4%

33.4%

63.7%

33.2%

20.4%

42.9%

6.1%10.5%

39.2%

26.4%

10.8% 13.2%

29.2%25.5% 22.7%

33.6%

7.6%

33.6%

18.3%

119.3%

34.5%

22.8%

71.8%

11.8%

22.3%

36.3%29.6%

19.5%24.3%

34.1%30.8% 32.9%

25.4%

200020012002200320042005200620072008200920102011

Significance for post-Keynesian Stock-flow consistent modeling

The complication of the models means that to actually simulate one, the modeler requires a set of equilibrium stock-flow norms that are attained in the steady state

of the model. Ideally, stylized facts should guide choices of stock-flow ratios. I’m here to tell you: they do not.

(Kinsella, 2011).

Significance for post-Keynesian Stock-flow consistent modeling

My research confirms also that companies funds investment mostly from current saving (net worth) and equities and there is no substitution between debt and

other sources of investment funding.