Using auctions to allocate goods and services is a common practice. Its popularity is understandable as auctions are often an efficient way to allocate goods to those with the highest willingness to pay. For example, one sixth of Sweden's GDP was allocated through public procurement auctions in 2017. In many of these auctions, several identical or similar goods are to be auctioned and it is common to do this sequentially. For example, this format is used to sell flowers, timber, wine and school milk and paving contracts but also by online auction houses (e.g. eBay and Tradera). In a sequential auction, goods should have the same price. For example, if there are lower prices in later auctions, it is worth waiting to buy. This prediction has been rejected in several studies that found lower prices in later auctions. The deposits constitute an anomaly called the "afternoon effect" and require further studies to clarify. The overall aim of this project is to investigate both empirically and experimentally what can explain declining price trends in sequential auctions. A first step is to understand how bidders' preferences can explain negative price trends. We also aim to understand how the auction format can explain the anomaly.