XVII. Working Cash - SoCalGas and SDG&E

Working cash is an allowance added to rate base that represents the funds provided by investors that are needed to pay for current operating expenses and provide a financial cushion. One major element in estimating the working cash need is the "lead/lag" that exists between when the company must pay its bills to suppliers for goods and services used to provide utility service and the time when the company receives payment for services from the consumers. This could be negative (in that the company needs cash to pay its suppliers ahead of receiving customer payment) or positive (customer payments arrive before payment to suppliers).

A. Federal & State Tax Payment Lag

ORA proposed to use data that it identified as 2002 tax payment information to determine the appropriate lag-days for SoCalGas and SDG&E.¹⁸⁴ Both companies disagreed.

SoCalGas argued that it had not yet filed its 2002 tax returns, but we observe tax payments do not necessarily follow the filing of corporate tax returns. In its reply brief, ORA points out that it had, in fact, had "used the actual 2002 fourth quarter payments for 2002, i.e. the actual check amounts paid for tax. The actual fourth quarter payment for 2002 was the stated payment of $56.974 million (federal tax) and $19.888 million (state tax) in Exhibit 81, p. 6, line 4¹⁸⁵" excluding the RTP Adjustment in Decmeber 2002. ORA's raises the concern that overpayments and refunds could allow SoCalGas to game the system. A quick review of SoCalGas' rebuttal testimony highlights ORA's concern.

The majority of the payments were made in the second half of the year for both 2001 and 2002. ORA points out that the lag days associated with the refund made in March 2003 for 2002 and December 2002 for 2001 is overstated since it is calculated from July of the previous year based on the assumption that the full payments were made on time rather than from the date of the overpayment. While the record does not provide enough information to calculate the actual lag days for refunds, ORA nonetheless raises a very valid concern that the company may not have calculated the lags associated with refunds correctly. Removing the refund adjustments, SoCalGas shows that for the three years 2000 through 2002, the federal lag days is between 78 and 87 days while the state lag days is between 67.4 and 70.0 days.

We will adopt ORA's Federal and State tax forecast of lag days for SoCalGas and SDG&E.

B. Municipal Taxes

TURN and UCAN argued that SoCalGas and SDG&E have understated the "lag-days" and they proposed to "synchronize" the taxes to the gas commodity prices.¹⁸⁶ SoCalGas and SDG&E responded that the lag days are not wrong and that the intervenors incorrectly use the (lower) commodity charge instead of (higher) total revenues - taxes are charged on total revenues.¹⁸⁷ For SoCalGas, a correct adjustment for revenue lag would be 42.61 days, based on a 13-month methodology. Included is a component for the collection lag that SoCalGas corrects to 25.40 days. By using both the correct base for taxes and the corrected lag days, SoCalGas reduces its rate base component for Municipal Taxes from $19.003 million to $15.277 million, a $3.727 million reduction.¹⁸⁸

We accept the SoCalGas corrections as the most reasonable estimation method and adopt the corrected calculation, above.

C. Revenue Lag

Revenue lag is the delay between customer receipt and payment of their utility bill. SoCalGas used 2001 base year to estimate its forecast revenue lag days of 42.73 days (25.52 days for collection lag and 17.21 days for other lags). In particular, SoCalGas used the 13-month period December 200 to December 2001 for accounts receivable and from January 2001

TURN recommends adopting a lower collection lag of 21.20 days based on the 12-month period October 2001 through September 2002. TURN claims that SoCalGas used a 13-month period which coincided with unusually high natural gas prices. Further, TURN points out that "in the eight months from through September 2001, receivables averaged 84.8% of SoCalGas' monthly sales. However, in the corresponding nine months from January through September 2002, receivables averaged only 71.18% of sales.¹⁸⁹" In contrast, SoCalGas argues that TURN has selectively used unusually low gas prices for each month of the time period used by TURN. Both claim that the other has used selective time periods to forecast the revenue lag.

It is important to use a time period which properly reflects reality. SoCalGas' initial testimony would forecast the revenue lag based on the test year 2001 - which it claims is the correct time frame as it's the same time period used as the basis for the revenue requirement request. While we have generally accepted the companies' forecast methodology which uses 2001 recorded as the basis, we also need to recognize when it is appropriate to deviate for legitimate reasons. In this case, the base year coincides with high natural gas prices. Indeed, other proceedings at the Commission are currently looking into this precise matter. It would not be wise to ignore more recent date in light of the unusually high prices during the 13-month time period used by SoCalGas. While it may or may not be true that TURN's time period has unusually low natural gas prices, it is certainly true that SoCalGas' time period has unusually high prices. We will adopt TURN's recommended forecast for revenue lag days.

D. Accrued Vacation and Withholdings

TURN and UCAN proposed to adjust the estimate for accrued vacation and withholding by using the adopted labor escalation rate. SoCalGas agreed (Ex. 27, p. 7) but disputed the second proposal that the change in accrual should reflect the same percentage change in O&M labor over the base-year. SoCalGas argued that "older workers" with 15-years or more of service earn more vacation (four to six weeks) compared to new replacement employees who earn only one to two weeks for the first five years. SoCalGas and SDG&E argued that as new employees are hired, they replace more experienced employees who are promoted. We do not find sufficient evidence to support the calculation and so we will not make the adjustment at this time.

E. Customer Deposits

SoCalGas and SDG&E tariffs require that an applicant for new service establish credit. A customer who has not qualified for the establishment of credit with the utility must submit a deposit pursuant to the utilities' Commission-approved tariff rules. SoCalGas and SDG&E are required to refund customer deposits with interest within 12 months as long as the customer pays his or her bills timely. Between 1997 and 2001, SoCalGas had average deposit balances of $35.088 million¹⁹⁰ and SDG&E had balances between $22 million and $24 million.¹⁹¹

TURN and UCAN contended that customer deposits represent a permanent source of working capital not provided by investors. However, customer deposits are not counted as an offset to the operational cash requirement under current practice. This is because SoCalGas and SDG&E pay interest on the deposits (at the three-month commercial paper rate, which is currently less than 2%), and the Commission's Standard Practice U-16 (U-16) indicates that while interest-free customer deposits should be counted as working cash, those on which the utility does pay interest should not be counted as working cash.¹⁹²

TURN and UCAN therefore recommended that the Commission, as a matter of policy, amend the "interest-free" restriction of U-16 as it specifically relates to customer deposits held by SoCalGas. TURN's specific recommendation in this proceeding was to reduce the test year rate base by $43.6 million based on escalating the 2001 data for TURN's forecast of customer growth to mid-2004. UCAN calculated $26.070 million for SDG&E. TURN and UCAN further recommended that if the Commission treats deposits as a rate base offset, it should also increase SoCalGas and SDG&E's operating expenses to reflect interest payable on customer deposits at a projected interest rate of 2%.

This recommendation is identical to TURN's recommendation in Edison's A.02-05-004 and adopted in D.04-07-022.¹⁹³

SoCalGas and SDG&E vigorously opposed the adjustment,¹⁹⁴ as Edison did in its proceeding. Applicants argued the recommendation conflicted with the standard practice, but for Edison the Commission found it reasonable to deviate from U-16. Edison lost its argument in A.02-050-004 that was based on comparing customer deposits to its historical investment in fuel oil inventories that was limited, for ratemaking purposes, to short-term interest. The Commission found:

"in rejecting SCE's proposal (in 1996) to rate base a portion of fuel inventory, the Commission held that "the risk Edison is offering to assume [of a change in value of the inventory] is not significant enough to justify a change in financing of the carrying costs." (64 CPUC 2d 241, 382, Finding of Fact 110-111.) SCE has not demonstrated to our satisfaction that the circumstances that led the Commission to reject SCE's proposal to rate base fuel inventory are equivalent to the circumstances attendant to TURN's proposal for customer deposits."¹⁹⁵

SoCalGas and SDG&E made essentially the same arguments as Edison. SoCalGas and SDG&E argued that they have large comparable amounts of shareholder capital committed to serving customers that does not earn a full cost of capital return, and the proposed treatment of customer deposits is inconsistent. SoCalGas and SDG&E argued that if the Commission treated customer deposits as an offset to rate base that would otherwise earn the authorized rate of return on rate base, to be fair we would have to change the treatment of other similar items that have previously been authorized to only earn interest at the short-term commercial paper rate. SoCalGas further stated that it currently has an average of $53.8 million of working gas inventory, where it is only authorized to earn the short-term commercial paper rate, not the authorized rate of return.¹⁹⁶ Applicants argued there is no reason to provide a lower return on this investment in inventory, provided by shareholders, than would be paid on the customer deposits provided by new customers. There is a comparable argument for SDG&E that has a $21.2 million investment in working gas inventory.

The question before us now is whether the current ratemaking treatment for SoCalGas and SDG&E is distinguishable from Edison.

The Commission has previously decided that the working gas inventories should accrue a short-term interest allowance.¹⁹⁷ These inventories are similar to Edison's fuel oil inventories in that they are both necessary for the operation of the system and they were found to lack the same element of risk associated with assets that earn the full rate of return when included in rate base, as was described in D.04-07-022. These are not the proceedings where SoCalGas and SDG&E may pursue a change in the ratemaking treatment for working gas inventories based on the equity argument or other arguments addressing the current risk compared to allowed return.

SoCalGas and SDG&E situation cannot be distinguished from Edison's and so we may consider applying the same treatment accorded to Edison's customer deposits to those of SoCalGas and SDG&E. Customer deposits are risk-free to SoCalGas and SDG&E, and the cost for the use of these funds is limited to the interest accrued and paid to customers when deposits are refunded.

If SoCalGas and SDG&E had no customer deposits, would they have to either borrow more money for working cash or to finance long-term plant included in rate base? Applicants argue the deposits do not finance long-term rate base because the funds are temporary.¹⁹⁸ Because the balance turns over continually as old deposits are refunded and new ones are made, we agree with SoCalGas and SDG&E that customer deposits do not finance long-term investments. But customer deposits can replace short-term borrowing. SoCalGas and SDG&E did not argue the customer deposits did not replace short-term borrowing. U-16 only calculates a needed amount of working cash¹⁹⁹ but the applicants have not shown they will borrow that amount as short-term debt in the test year. Applicants' ratemaking models do not show customer deposits as available funds not provided by investors. We can conclude that in practice, SoCalGas and SDG&E use the deposits in lieu of borrowing the amount projected by U-16.

Therefore, we will adopt the recommendations of TURN and UCAN and reduce the working cash rate base component by the average of $35 million and $23 million for SoCalGas and SDG&E, respectively. We will authorize SoCalGas and SDG&E to recover interest at the forecast commercial paper rate of 2%, or $0.700 million and $0.460 million, respectively, in the 2004 base margin.

F. Hub and Swap Revenues

In its opening litigation brief,²⁰⁰ TURN proposed to reduce by 25% the Accounts Receivables attributable to hub and swap transactions, or $2.6 million, that would reduce the working cash calculation for SoCalGas. TURN argued that hub and swap transactions utilize core capacity and commodity assets and therefore working cash requirements should be reduced because revenues generated by hub transactions offset the purchased cost of gas and contribute to savings that may result in increased shareholder rewards under the Gas Cost Incentive Mechanism (GCIM).

SoCalGas responded²⁰¹ to this proposal by pointing out that the revenues are recorded in full in the Purchased Gas Account (PGA), which has the effect of reducing gas costs recovered from core customers. Subsequently, under the GCIM there is sharing incentive mechanism that allocates any gains outside of a deadband between customers and shareholders. SoCalGas argued that this proposal is a piecemeal change to the GCIM.

We disagree. TURN correctly points out that "without the use of core assets for transactions it is much less likely that shareholders would see any benefits.²⁰²" We will adopt a reduction to Accounts Receivable by 25% to reflect hub and swap transactions as proposed by TURN.

¹⁸⁴ Ex. 301, p. 26-5 and Ex. 302, p. 26-5.

¹⁸⁵ ORA reply brief, p. 12

¹⁸⁶ Ex. 501, p. 49.

¹⁸⁷ Sempra opening litigation brief, p. 286.

¹⁸⁸ Ex. 127, pp SWK-3 through SWK-7, and SWK-13.

¹⁸⁹ TURN opening brief, p. 154

¹⁹⁰ Ex. 501, p. 52, excluding 2002, which is beyond the base year.

¹⁹¹ Interpreting a chart without a data table, see Ex. 602, p. 75.

¹⁹² Ex. 501, pp. 51-53.

¹⁹³ See D.04-07-022, mimeo., pp. 251-257.

¹⁹⁴ Sempra opening litigation brief, pp. 292-294.

¹⁹⁵ D.04-07-022, mimeo., pp. 256-257.

¹⁹⁶ Ex. 127, p. SWK-9.

¹⁹⁷ Sempra opening litigation brief, p. 293.

¹⁹⁸ Sempra opening litigation brief, p. 293.

¹⁹⁹ It is possible that working cash would exceed needs, but the applicants' calculations show a need for additional working cash that becomes the rate base component in dispute.

²⁰⁰ Pages 147-148.

²⁰¹ Sempra opening litigation brief, p. 295.

²⁰² TURN opening brief, p. 148