We are interested in computation of locating arbitrage in-nancial markets with frictions. We consider a model with a nitenumber of nancial assets and a nitenumber of possible states of nature. We derive a negative result on computational complexity of arbitrage in the case when securities are traded in integer numbers of shares and with a maximum amount of shares that can be bought for a xed price (asin reality). When these conditions are relaxed, we show that polynomial time algorithms can be obtained by applying linear programming tech-niques. We also establish the equivalence for no-arbitrage condition & optimal consumption portfolio.