Subject: Stocks and Flows

Posted by David Wheat on 3/3/2004
In Reply To:Stocks and Flows Posted by Alfred Bosshard on 3/3/2004

Message:

Good point, Alfred, about “point of view.” GDP is an example.

Real-time GDP is technically a flow, and the units are dollars per year. However, economic policy-makers perceive a “statistic” called GDP, which is a piece of information about recent average production of goods & services. That statistic is a stock (still measured in dollars/year). The policy-makers are primarily interested in the change in the GDP; i.e., whether it’s accelerating or decelerating, and by how much. In that case, the “flow” of interest is the change in GDP, measured as dollars per year per year.

A model should include both perspectives. GDP would be modeled as a real-time flow of goods & services. The information about GDP (“perceived GDP”) would be modeled as a stock.

As far as teaching goes, the first step is to convey the real-time sense of things (e.g., GDP is a flow), and only later muddy the water by explaining that the perceived GDP (the government-collected statistic) is an information stock.

Lees’ example of deficits and debt can be treated similarly. Clearly, she’s right that the deficit (or surplus) is the flow that changes the stock of debt. However, policy-makers may focus on the deficit per se, in which case they collect statistics about recent average deficits or project future average deficits. In that case, it is the information about the deficit (recent average or projected future, both measured in dollars per year) that would be the “condition” that is acceptable or unacceptable; i.e., an information stock that is meeting its goal or not. The change in the deficit would be the flow variable of interest. Again, a model should include both formulations—deficit as a real-time flow, and “perceived deficit” as an information stock.

Dave