The Amsterdam office investment market is not fully efficient. It is slow to incorporate essential new information. The main problem of the Amsterdam office market is long-term vacancy clearly higher than natural vacancy. Total vacancy soared from 250,000 m2 in 1998/1999 to 1.5m m2 in 2004. Recently, vacancy moved around 1.6m m2. This represents almost 20% of the office stock. In many suburban areas vacancy is 25% or higher. Nevertheless, construction of new office is continuing as there is steady demand for modern office space. In fact, vacancy of newly built offices is low, while the vacancy of older offices is rising to unprecedented levels. It proves that new construction gives rise to a filtering down process. Older offices are often difficult to let on conventional lease terms. New offices can move to the secondary market, sometimes after only five to ten years.Interestingly, this risk for office investors is not reflected in investment yields. Since the end of the 1990's, reasonable expectations have changed completely, but yields in Amsterdam remain lower than in elsewhere in the Randstad area. Investors should consider that their properties will only for a relatively short term belong to the prime office market. However, it is fully unclear how they value this risk. We postulate that this remarkable situation can be explained by the fact that most capital for office investment is supplied indirectly. There are two main sources of capital: German open-end funds and Dutch private partnerships. Retail investors subscribe to these funds and partnerships without fully understanding the risks of real estate investment. Subsequently, the funds are under pressure to invest their liquidities. If our analysis is correct, a solution is near. Several German open-end funds will be liquidated and many Dutch partnerships have serious problems. The amount of money available for investment in offices in Amsterdam is falling and as a consequence, yields will rise to a level that better reflects the risks.