EXHIBIT 99.488

                    THE EFFECT OF LARGE SUPPLY WITHDRAWAL ON
                            THE MARKET CLEARING PRICE


When a large supply into a zone is withdrawn for one reason or another, what is
the effect on the zonal market clearing price? The answer to such a question has
real implications to the behavior of California energy market, as we will later
discuss on the case of the PX Day-Ahead energy market for the dispatch day of
May 25, 1999. To approach the answer to this question, let's first examine the
following simple example.


THE EFFECT OF WITHDRAWING A SUPPLY PORTFOLIO BID

As shown in Figure 1, Zone 1 in this example has one internal supply portfolio,
SPA, one external supply, SPB, and one demand portfolio DP. The portfolio bid
curves are shown in Figure 2. Also shown in Figure 2 is the aggregate supply bid
curve. The intersection point of the aggregate supply bid curve and the demand
bid curve gives the market clearing price (MCP) and clearing quantities.


                                    [CHART]

                     Figure 1: A Zone with External Supply






                                    [CHART]

                         Figure 2. Portfolio Bid Curves

            Figure 2 provides several simple and useful observations.

1.   The aggregate supply bid curve is generally less steeper than any of the
     composing individual supply bid curves. Same can be said of the aggregate
     demand bid curve, although in this simple example the aggregate demand bid
     curve is the only individual demand bid curve.

2.   Withdrawing the supply portfolio bid, SPB, from the auction will reduce the
     aggregate supply bid curve to the bid curve of the supply portfolio bid,
     SPA, which is steeper. In fact, Figure 2 shows that withdrawing SPB from
     the auction will raise the market clearing price from the value MCP to the
     value MCP'.

In general, withdrawing any supply portfolio bid from the energy auction will
raise the unconstrained market clearing price.


THE EFFECT OF WITHDRAWING A SUPPLY RESOURCE SCHEDULE

What if SPB is not withdrawn from the auction but its composing resource
schedules is withdrawn later in the ISO congestion management process due, for
example, to a inter-zonal interface outage from Zone 2 to Zone 1 shortly before
the ISO congestion management process?





To answer this question, we first make two assumptions that will help simplify
the subsequent discussion without limiting the general conclusions. First, we
assume that each portfolio schedule is fulfilled by one resource schedule.
Therefore, we have supply resource schedule SRA for portfolio SPA, supply
resource schedule SRB for portfolio SPB, and demand resource schedule, DR.
Second, we assume that the adjustment bids of the resource schedules are
conforming, or otherwise consistent with the portfolio bid curves. By being
consistent with the portfolio bid curves we mean that the adjustment bid curves
more or less wrap around the portfolio bid curves, as illustrated in Figure 3.


                                    [CHART]

                        Figure 3. IPS and Adjustment Bids

With the above two assumptions, we are ready to examine the effect of
withdrawing the resource schedule, SRB (whose portfolio bid is SPB) on the
market clearing price in Zone 1.
Referring to Figure 3, since SRB is withdrawn, we are left with only two
adjustment bid curves, that of the supply resource SRA and that of the demand
resource DR. The intersection of these two adjustment bid curves yields the new
market clearing price and quantities. It can be observed from Figure 3, this
intersection point almost coincide with the intersection point of the supply
portfolio bid curve of SPA and the demand portfolio bid curve of DP. For this
example, therefore, we conclude that the effect of withdrawing SRB, the resource
schedule composing of the portfolio bid, SPB, is very similar to withdrawing the
portfolio bid, SPB, itself in the auction. The market clearing price in Zone 1
resulting from withdrawing the resource schedule, SRB, in congestion management
is very close to MCP'.





In general, we can deduce that, as long as the adjustment bids are conforming
(or to reflect the price-quantity relationship of the originating portfolio bid)
the effect of withdrawing all resource schedules composing of a portfolio
schedule on the market clearing price is similar to that of withdrawing the
original portfolio bid.


THE EFFECT OF NON-CONFORMING ADJUSTMENT BIDS

What if the adjustment bid is not, or cannot be, conforming so as to reflect the
portfolio bid's price-quantity relationship? Let's continue our discussion by
modifying our example. Suppose, that the demand portfolio, DP, is fulfilled with
an inter-SC resource schedule, DR. As a result, DR does not have an adjustment
bid, or we can say that DR's adjustment bid is now as shown in Figure 4.


                                    [CHART]

                   Figure 4. IPS and Non-conforming Adjustment


As evident in Figure 4, withdrawing the resource schedule SRB now will move the
market clearing price to MCP", considerably higher than MCP'.

In general, non-conforming adjustment bids will distort the market clearing
prices. In the case of inter-SC trade, the present inability to accept
adjustment bids for inter-SC trade resource schedules can lead to considerably
higher market clearing prices when a large supply resource schedule is withdrawn
in congestion management, as illustrated in the example depicted in Figure 4.





THE EXAMPLE OF MAY 25, 1999 PX DAY-AHEAD ENERGY MARKET

What happened to the PX Day-Ahead Energy Market for dispatch day of May 25, 1999
is analogous to the simple example presented in Figures 1 through 4. For the
dispatch day in question, the composing supply resource schedules for a 2900 MWh
supply portfolio schedule external to California were withdrawn due to
transmission constraint, causing a sharp rise of the market clearing price
within California. In this case, California is analogous to Zone 1 in the simple
example of Figure 1 and the 2900 MWh supply portfolio external to California is
analogous to SPB in the simple example of Figure 1. Figure 5 provides the actual
aggregate demand portfolio bid curve and the actual aggregate supply portfolio
bid curves for hour ending 15 of May 25, 1999. Also shown on Figure 5 is the
aggregate supply portfolio bid curve without the 2900 MWh supply portfolio bid.

                                    [CHART]


             FIGURE 5. Aggregation Portfolio Bids for Hour Ending 15
                      of the May 25, 1999 Day-Ahead Market


Following the same reasoning as we have used for our simple example previously,
we can see that withdrawing the 2900 MWh supply resource schedules into
California has the natural effect of raising the California market clearing
price from nearly $29/MWh to nearly $36/MWh, for hour ending 15 of the May 25,
1999 Day-Ahead market, assuming





that all resource schedules are accompanied by conforming adjustment bids. In
reality, however, the adjustment bids do not coincide very well with the
portfolio bids, in other words, they can be highly non-conforming, as can be
observed from Figure 6, where the aggregate supply adjustment bid curves and
aggregate demand adjust bid curves are shown for hour ending 15 of the May 25,
1999 Day-Ahead market. The intersection of the aggregate supply and demand
adjustment bid curves yields the actual PX Day-Ahead zonal market clearing price
in California for hour ending 15, $48.98/MWh.


                                    [CHART]


  FIGURE 6. Aggregate Adjustment Bid Curves and Aggregate Portfolio Bid Curves
             For Hour Ending 15 of the May 25, 1999 Day-Ahead Market


It is also instructive to observe that about 7600 MWh of inter-SC supply
portfolio bids and about 2800 MWh of inter-SC demand portfolio bids were
scheduled for hour ending 15 of the May 25, 1999 Day-Ahead market. Almost all of
these inter-SC portfolio bids were price takers in the auction. None of the
inter-SC resource schedules could contribute adjustment bids to the congestion
management. If such inter-SC resource schedules are allowed to participate in
congestion management, the composite adjustment curves would have likely been
less steeper, leading to lower market clearing prices after congestion
management.