A theory is developed relating the international money movement methods used by mult i national companies to the cash flow patterns Generated by their bu s iness. Des pite the general int r est in this controversia l area , a s urvey of the managerial lit erat ure indicated that it was und rde v Jloped and with a strong bia s to prescriptive mod els that were accompani ed by little underlying research . A de tailed manageria l survey was th erefore carri ed out by questionnaire and interview; this showed that the mod I s proposed by the literature were not being used but that several simple ' techniques ' were be ing exte nsively applied. The th eory was th n developed to explain the circums tances in which each of the ' technique s ' were used. The questionnaire data was th en analysed using three di ffe rent statistical techniques to see whether the companies ' practice in fact corresponded with the theory's predictions. lThe statistical techniques were canonical correlation, stepwise multiple linea r regression , and discriminant analysis. ) The invest igations showed considerable a r eas of agreement between the th eory developed and current practi ce with few unexplained results . The statistieal r esults were the n further interpreted to generat a set of practical decision rules for manager s . These rules use data about a company ' s international ca sh f l ows to indicate the money management systeln that would typically be appli ed to those cash flows. Two mathemacital models for th e optimisation of cash holdings nnd tra nsfers wer e also built a nd discu s s ed, One model cOllside r ed the ce ntralisa t i on of cash r ese rves , a nd the otller the optima l planning of tra nsfers after mult'natinnal ne tting techniques have been a pplied, Doth models are develope d within a contex t of des iring to promot e d e ci s io ll rules which implicitly s tructure cash flows for the be ne fit of a ll the countries affe cted by a mul tinationa l company ,