Paradoxically Rejected Block Order in a Day-Ahead Auction
In a single-hour day-ahead auction, demand is 850 MW plus one block bid from RetailCo: buy 100 MW at a limit price of €75/MWh, all-or-nothing. Supply offers are G1 800 MW at €40/MWh, G2 100 MW at €60/MWh, and G3 100 MW at €80/MWh. G2 has a minimum run time of two hours and a start-up cost of €4,000, but it is already committed and its start-up cost is sunk for this hour. G3 can start in the hour. The exchange uses marginal pricing and accepts a block order only if it is in the money at the resulting clearing price; there is no make-whole payment for rejected blocks. The price cap is €300/MWh, and no imports are available. The market operator must decide whether the outcome is efficient and whether a rule change is justified.
1. Clear the auction without the block and with the block, state whether the block is accepted, and explain the paradox.
Read the worked answer
Without the block, demand is 850 MW. G1 supplies 800 MW at €40, and G2 supplies 50 MW at €60. G3 is not needed. The clearing price is €60/MWh.
With the block, demand is 950 MW. G1 supplies 800 MW, G2 supplies 100 MW, and G3 supplies 50 MW at €80. The clearing price is €80/MWh. At €80, RetailCo's block bid of €75 is out of the money: if accepted, it would pay €80/MWh, above its limit. The all-or-nothing rule therefore rejects the block. But once rejected, the price falls to €60, at which the €75 block would have been in the money. This is the paradox: the block is profitable at the price that prevails when it is rejected, but accepting it raises the price above its bid.
The incremental cost of serving the block is 50 MW from G2 at €60 plus 50 MW from G3 at €80 = €7,000. Its value is 100×75 = €7,500. Acceptance would therefore raise social welfare by €500, yet the uniform-price rule without make-whole rejects it.
2. Assess whether the market operator should introduce a make-whole payment or another rule to accept this block. Consider efficiency, strategic bidding, and uplift allocation.
Read the worked answer
The €500 welfare gain is small but real. A make-whole payment of up to €500 to RetailCo could allow acceptance while ensuring the block pays no more than €75/MWh. A pay-as-bid rule for block orders could also solve the paradox, but it may distort bidding for other orders. The operator should compare the administrative cost and strategic risk against the frequency of such blocks.
Strategic risk is significant. If load knows it can receive a make-whole payment, it may understate its limit price to extract compensation. The operator should require the block's bid to be firm, monitor repeated claims, and cap compensation at the true incremental cost. The G2 start-up cost is sunk here; if it were not sunk, incremental cost would be €11,000, above the €7,500 value, and rejection would be efficient. The rule should therefore use the actual incremental cost for the hour, not a blanket make-whole.
Any uplift should be allocated broadly to demand rather than to the block itself, so that the block's bid remains the primary signal. The operator should also publish the frequency and cost of paradoxically rejected blocks to justify any rule change.