'Macroprudential surveillance' - monitoring conjunctural and structural trends in financial markets so as to give warning of the approach of financial instability - is immensely important, given that financial crises can have huge costs. In this context, this paper presents three complementary lectures, which set out in generic terms the financial data needed for monitoring risks of financial instability. The paper starts with a view of the nature of financial instability, and the types of turbulence, that might pose particular systemic dangers, and the implications they have for data needs. These together give building-blocks for the listing in the third lecture of the types of financial and macroeconomic data that are needed for macroprudential analysis, and a suggested approach to their interpretation. A practical example is given, by looking at how theory and data respectively gave clues to the approach of the Asian crisis of 1997-8, and in this context, notes the data actually available for Thailand at the onset of the crisis in 1997. Overall, it is suggested that the theory of financial instability and the experience of financial crises in the past provide sufficient material to enable meaningful use to be made of financial and macroeconomic data in macroprudential surveillance. Such data may include econometric forecasts, as well as current information. In using such data, judgement is crucial in assessing risks to financial stability - macroprudential surveillance can never be mechanistic. Nevertheless, the paper maintains that detailed knowledge of the sequence of events in past crises, both directly and as encapsulated in theory, is a sine qua non to interpreting the data. In addition, there is a need for development of broad information on what constitutes normal conditions in an economy, as well as the patterns that have often preceded financial crises in the past both domestically and internationally. Given the shortcomings in the data available for many countries, especially in the emerging markets, considerable efforts to improve coverage and timeliness are warranted. Besides macroeconomic data, emerging-market countries may need to lay particular emphasis on better banking data, given the structure of their financial markets, which is typically bank-dominated. Private sector agents also have a role to play in monitoring the risks they face as a consequence of the behaviour of the overall financial system. They are, therefore, encouraged to undertake their own analyses of risks at a macro level.
