Abstract:

In a bilateral supply chain， the retailer sources product from the supplier to satisfy the random demand. The supplier needs to produce the product before the demand is realized. The supplier is financially constrained and can borrow money from the bank if necessary. As the capital market is perfectly competitive， the bank will determine the interest rate based on the loan amount and the associated risk so as to maintain its expected return rate equal to the exogenous risk-free rate. In order to reduce the financing cost of the supplier and induce him to produce more product， the retailer has a motivation to provide a （partial or full） guarantee for the supplier’s loan. A key feature of our model is that the guarantee is not a hard constraint of the loan amount， i.e.， the supplier can borrow as much as he needs， regardless of the retailer’s guarantee proportion. The retailer and supplier play a Stackelberg game. The retailer needs to decide the purchasing price and guarantee proportion as a leader， and then the supplier determines the production quantity as a follower. Both the retailer and supplier are risk-neutral profit maximizers. By backward induction， the supplier’s decision is studied， and then retailer’s problem is considered.For any given purchasing price and guarantee proportion， the supplier’s expected profit is quasi-concave in the production quantity， and thus his optimal-response production quantity is unique. The retailer’s optimal policy is to provide a full guarantee for the supplier’s loan. The retailer’s guarantee will not change he financing cost of the whole supply chain， but can shares the supplier’s financing cost. By providing a full guarantee， the retailer can maintain the supplier’s production quantity unchanged while offering a lower purchasing price. The overall effect of full guarantee is beneficial to the retailer. In other words， a full guarantee causes larger financial loss but brings higher sales profit to the retailer. The retailer’s optimal expected profit is demonstrated to be decreasing in the supplier’s internal capital. Thus， the retailer will be better off to work with a less wealthy supplier. Finally， the bankruptcy cost is incorporated into the model. It turns out that the presence of bankruptcy cost is unbeneficial to the retailer， but will not change the retailer’s guarantee policy.

Key words: capital constraint, supplier finance, buyer guarantee, limited guarantee proportion, bankruptcy cost