Extrinsic Uncertainty typically exacerbates the beliefs of the general public –the depositors in this case— without it being associated with any change in the fundamentals of the economy. This type of shock is frequently linked with outcomes of a self-fulfilling type: a prophecy which outcome becomes true through the beliefs and actions of the main economic actors. The literature frequently illustrates the properties of this type of shock by using the example of all of the young domestic depositors having a “bad dream” in which banks will close, but without any change to the fundamentals of the economy. Interestingly, there is a coordination problem in our model economy in which complementarity is present in the strategic interaction between the individual depositors: if individual households cannot observe the actions or types of others, they may panic if they believe that every- one in the economy has had the same bad dream. In other words, if an individual depositor believes that every- body else is going to run and banks have a sequential service constraint in place, she will run as well, and, in the
31
aggregate, all depositors will run on banks19. In particular, whether depositors choose to run or not will be deter- mined by the results they obtain from the sequential checking mechanism, which is shown in Figure 6. This me- chanism is informationally efficient and it is defined as the algorithm followed by domestic depositors, in which they check several conditions related to the banks’ positions. In the event of a shock, banks do not get a chance to re-optimize, since there is no change in the fundamentals of the economy, but all domestic depositors may run on the banks to withdraw their resources immediately --aiming at being the first in line-- and the financial system would have a fully blown self-fulfilling attack on banks in its hands. As we will see later in detail, there exists a particular set of circumstances under which this “bad dream” has the potential of greatly affecting the outcomes in the economy by leading to panic-equilibria with significantly lower welfare.
Intrinsic uncertainty is the second type of uncertainty that we explore in this paper. It has to do with an unan- ticipated change in one or more of the fundamentals of the economy after households have formulated their con- tingent plans and learned their true types. A sudden stop certainly belongs in this category, and we try to replicate it in this paper by an unanticipated and exogenous reduction of the amount of new credit available from the rest of the world ( f1>0.) Recall that the pair
{
f0,f1}
was set exogenously in the rest of the world at the beginning of period t, and that all contingent plans were made at that point. The two constraints on foreign credit were binding and banks had already acted on their choices of d0,t and d2,t+1, but not on their choices of d1,t+1 and lt. Moreover, the deposit contract{
c1,t,c2,t+1}
cannot be modified. When a sudden stop hits the economy, it abruptly reduces the resources available at the end of period t tof
1′
goods, where 0< f0< f1′< f1 obtains. The relevantforeign-credit constraint now becomes
2,t1 1,t 1 1
d ++d% + =f′, (32)
where d%1,t+1 denotes the re-optimized value of d1,t+1, since the latter is not feasible after the shock. We will see later in this section that this intrinsic shock may reduce both liquidity and solvency in the financial sector20, thus having the potential of triggering a domestic crisis that may take the form of generalized attacks on banks, bank- ruptcy and closure in the domestic financial system.
Once such a shock hits the economy, this event becomes public information and the availability of future resources changes irreversibly. Under this new set of circumstances, it may be in the depositors’ best interest to withdraw from banks as much as possible immediately, and the presence of a sequential service constraint may only exacerbate this problem. In these circumstances, it may not be optimal for any agent –specially, households of the patient type-- to wait until the next period to withdraw from the banks, since they believe that banks could be facing bankruptcy and closure. Both banks and depositors will need to re-optimize to account for the change in circumstances, leading again to the Sequential Checking Mechanism. This algorithm consists of three steps. The
19
The reader may also be interested in suspension of convertibility as a measure that could preclude a bank run from happening. However, we only focus in the simplest case here.
20
One could also argue that unanticipated reductions in foreign credit may trigger a shock to the preferences of depositors. If such a shock induces a crisis of a self-fulfilling nature, this may only exacerbate the existing problems in this economy. In this paper, for simplicity, we abstract from this possibility.
32
first step is an evaluation of the liquidity position of banks. Next, evaluating the banks’ solvency becomes the next priority, followed by checking whether the resulting allocations are incentive-compatible or not.