Rating it yourself: Kaplan-Urwitz model

Ever wondered how rating companies like Standard and Poor or Moody`s issue credit rating? These credit rating companies use a combination of statistics and financial ratios to arrive at a credit score.

Over time, researchers have tried to predict the formula used by these companies and one of the models used in financial management is the Kaplan-Urwitz model. There are separate formulae for quoted and unquoted companies:

1) Quoted companies 

The formula for quoted companies is given as:

Y = 5.67 + 0.011F + 5.13π - 2.36S – 2.85L + 0.007C – 0.87β – 2.90σ

Where

Y = The score the model produces

F = Size of the firm (Total assets) 

π = Net income / total assets 

S = Debt status ( subordinated debt = 1, else = 0)

L = Gearing (Long term debt / total assets)

C = Interest cover (PBIT / Interest payment)

β = Beta of the company from CAPM 

σ = Variance of the residuals from CAPM equation

 ( √(σ² – β² – σ²_m ) )

where σ²_m is the variance of the market

 2) Unquoted companies 

There is a subtle difference in the constants of the equation and no beta is present given this is a company that is unquoted. 

Y = 4.41 + 0.0014F + 6.4π - 2.56S – 2.72L + 0.006C– 0.56σ

 Where (Only σ is different from the top)

Y = The score the model produces

F = Size of the firm (Total assets) 

π = Net income / total assets 

S = Debt status ( subordinated debt = 1, else = 0)

L = Gearing (Long term debt / total assets)

C = Interest cover (PBIT / Interest payment)

σ = Standard deviation of earnings


Credit rating categories 

Score (Y)	Rating
Y > 6.76	AAA
Y = 5.20 – 6.75	AA
Y = 3.29 – 5.19	A
Y = 1.58 – 3.28	BBB
Y = 0 – 1.57	BB

For those with Harvard log in credentials, the full research can be obtained here.

Reference:
1. ACCA P4: Advanced Financial Management. London: BPP, 2013. 230-32. Print