Enron Case Study

Going by the effect of change in temperature in the past four years on the revenues of PNW, we know that a hedge against weather risk is of paramount importance for Mary Watts, the chief financial officer of PNW. The net income of PNW for 1999 was only 8 million while the yearly capital expenditure from 1995-1999 was 1 billion. Under this red ink budget, it came as no surprise when the equity and debts holders increased their required compensation of risk. This high cost of capital further worsened the future EPS prospects of PNW. With the forecast for two warm winters, PNW would be better off reducing their exposure to weather risks by considering the weather derivatives.

Forecast of HDD

The correlation of the electricity demand to the change of HDD is about 91.7%. The demand of electricity also represents the profits PNW can earn. In order to hedge the loss that unstable weather may bring, PNW could buy weather derivative contracts. If we could accurately predict the HDD next year, then we would know which contract of weather derivatives Enron offers would be the best fit for PNW. In this part we apply the statistical method of 95% confidence level to find out the possible range that future HDD may fall within. We then compare the HDD confidence level with the strike amount that these contracts offer.

Before calculation of the 95% confidence level of HDD, one critical factor to be considered is adjustment for the cyclical effects caused by EI Nino, La Nina, or Heat Island. The goal here is to forecast the most accurate HDD for next year. We see that the last EI Nino effect happened in 1998, implying it won't recur until 2005 which we base upon its frequency of occurring every seven years. We should get rid of this special cyclical effect while forecasting the HDD in 2001 which is supposed to be a normal year without any cyclical weather effect. The data in the appendix is the HDD computed after the temperature was adjusted for increased trends and the volume was adjusted for increased customers. Below are the forecast procedures we used to arrive at the HDD for the next year:

1. Adjust the cyclical effect: Omit the data in 1992 and 1998.

2. Compute the mean and standard deviation after adjustments: They are 3,137 and 127.57.

3. Build a 95% confidence level:

3,137 ± 127.57 / √9 === > (3,053.66, 3,220.34)

4. Compare the forecasted HDD with the weather derivatives contracts.

The relationship between Volume and HDD

Run the regression of change in volume and change of HDD. Analyze the profits PNW lose and the risk compensation per HDD. Margin profit per MWb is 40 dollars.

Volume Change in Volume HDD Change in HDD

8825019 2978

9077284 252265 3207 229

8490052 -587232 2665 -542

9190089 700037 3410 745

9018775 -171314 3156 -254

8970078 -48697 3037 -119

8994810 24732 3032 -5

9146973 152163 3212 180

8962612 -184361 2892 -320

8998800 36188 3105 213

8989312 -9488 3094 -11

Using E-Views we calculate the regression:

Dependent Variable: Change in Volume

Method: Least Squares

Sample (adjusted): 2 11

Included observations: 10 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 6028.701 26256.26 0.229610 0.8242

Change in HDD 896.6034 77.08367 11.63156 0.0000

R-squared 0.944170 Mean dependent var 16429.30

Adjusted R-squared 0.937192 S.D. dependent var 331110.1

S.E. of regression 82981.43 Akaike info criterion 25.66748

Sum squared resid 5.51E+10 Schwarz criterion 25.72799

Log likelihood -126.3374 F-statistic 135.2932

Durbin-Watson stat 2.640235 Prob(F-statistic) 0.000003

Analysis of the Enron's weather derivatives

(1) Basic Floor

Breakeven: 1,338,000 / 35,750 = 37.43

Cap: 14,588,000 / 35,750 = 408.06

Strike Amount: 2,925 HDD

(2,925 - HDD) * 35,750

2,925

HDD - HDD

2,887.57 2,516.94

(2) Reduced Tick Size

Breakeven: 1,253,641 / 33,500 = 37.42

...