Task 1: Indicator winners and bidders

[MariaArnalCanudo]

New winners vs old winners (whether a firm has won a contract for the first time or not). (At contract level)

• Firms than have increased the products code during COVID (selling more than one product)
• Calculate the probability a firm had won a contract in the last X years based on the number of contracts won during COVID

[serenacocciolo]

Indicators to study the characteristics of the winners (and of the buyers):

• SME dummy [Are contracts during covid less likely to be awarded to SMEs? Are bidders during covid less likely to be SMEs?]

• Same location dummy: Whether the firm is from the same municipality, or from the same province or from the same state (we can look at different levels of geographical areas, depending on the country). [Are contracts during covid more likely to be awarded to firms from the same location as the buyer? Are bidders during covid less likely to be from the same location as the buyer?]

• New winner and new bidder: Reduce the contract-level data at month-year level. For each contract, calculate the number of months from the previous contract by the same firm (and assign 1,000,000 if the firm does not have any previous contract). We can use this "duration" variable (time btw contracts) to define new winners. For example, if the previous contract by that firm is more than X months before. We can calculate this indicator overall (ie new winner in public procurement overall) and by sector (ie. new winner in sector Y). [Are contracts during covid less likely to be awarded to new winners? Are contracts during covid less likely to be awarded to new bidders? Are bidders during covid less likely to be new bidders?]

• Number of different products (CPV codes) supplied by the same firm over 1 year (or 6 months). [Do firms during covid supply a broader range of different products?]