Bangko Sentral ng Pilipinas Deputy Governor Diwa C. Guinigundo stated recently that the private-sector analysts have triggered spiraling inflationary expectations last year, causing the BSP to tighten monetary policy by as much as 175 basis points. This statement comes after Finance Secretary Carlos G. Dominguez III announced early this year that private financial analysts will be made accountable for the accuracy of their macroeconomic forecasts, particularly on inflation. In turn, these pronouncements are derived from the recent study of the Department of Finance (DOF) that the inflation forecasts done by analysts of well-known institutions showed absolute deviations to be as high as 0.4 percentage points, amounting to an estimated margin of error of around 15 percent.
According to the DOF, based on international experience, forecasts partially contribute to higher than expected inflation as these created expectations resulting in “self-fulfilling prophecies.” While he has no intention of blaming these analysts, Dominguez warned that a similar review will be made at the end of the year.
To understand the consequences of this rule, it is important to know how forecasts are done. The art or the science of forecasting would begin by looking at historical data to see which factors and which policies significantly affect the economy. The development of econometrics naturally brought about an influx of macroeconomic models that tried to predict in the short and medium terms of the growth of gross domestic product (GDP), employment and inflation, among other aggregate variables. Unlike the studies in the field of microeconomics, macroeconomic forecasts analyze aggregate variables which are viewed to be more stable. A series of tests and simulations are then conducted before the results are usually released. Mainly on the basis of strong quantitative methods that underpin these forecasts, policy decisions are often made in light of these studies.
However, the value of forecasts has been severely diminished by the so-called Lucas’s critique. Written by Economics Nobel Laureate Robert Lucas, the analysis claims that forecasting the effects of economic policies entirely on the basis of relationships observed in historical data is wholly simplistic. The macroeconomic models that use structural equations that are invariant to changes in economic policies cannot be a reliable assumption for forecasting. The problem is that economic policies create their own expectations and, in the process, could change the way these variables, especially the highly aggregated variables, affect the other variables.
The critique then suggests that the validity of macroeconomic forecasting models depends on how well it can incorporate the people’s responses to policy shifts, and more important the “rational” expectations formed by such policies. It is the formulation of these expectations that will determine whether the model will be able to predict accurately the movements of GDP and inflation. Unfortunately, this is easier said than done since these will involve incorporating microeconomic foundations, which are beyond the scope of these forecasting models.
More important, the critique turns the DOF argument on its head. Forecasts do not affect the expectations. Rather, expectations influence the forecasts. Assuming that all models use similar (though not identical) structural equations that describe the economy, off the mark predictions conceivably have failed to anticipate correctly those expectations. The better forecasts will be those that capture the correct expectations.
Nevertheless, one has to consider that when these forecasts were made last year, we were in the midst of the highest inflation rate in a decade. National Economic and Development Authority Director General Ernesto M. Pernia noted that he was surprised by this event. Budget Secretary Benjamin E. Diokno also mentioned that we should get used to the high inflation and not behave like crybabies, indicating that we should be grateful that the inflation was lower than the historically high records. All these reactions created an atmosphere of desperation and uncertainty. Given the weak response from the government at the time, many of these forecasts unsurprisingly predicted the continued increase in inflation.
Conditions began to stabilize only when the government took matters in its hands. While the decline in petroleum prices is significant, what really contributed to the fall in inflation was the government’s action on the supply constraints, especially in agriculture. Thus, when the supply of rice started to enter the markets, a perceptible decrease in inflation was noted, and expectations of a lower inflation were formed. The point is that the government has a substantial influence in the way expectations are created. Policy reforms that adhere to efficiency rules, and not to sudden, inconsistent discretionary measures, are clearly the right tools toward conditioning these expectations.
Given the government’s hold on the economy, the warning to analysts to be accountable is constricting these forecasting institutions to conform always to the government prescriptions. When events deviate from the expected as they did last year, the government can use all of its power and resources to bring it back to where it wants to move. As seen from the DOF study, analysts who placed a premium on government claims and recognized its ability to manage the economy were able to achieve a lower margin of error. However, this may not always be the case. If petroleum prices did not decrease substantially toward the end of the year, the effects of the government’s anti-inflation plan would not have been as effective. In this case, analysts who took a more independent stand from the government estimates would have had more precise predictions. However, even if this was the case, those who would pattern their forecasts with the official accounts would not really be worse off since the mistake is shared with the government.
If the government intends to eliminate the negative impact of expectations, “scoring” these private forecasting institutions will not be the right policy. Instead, the government should take a more consultative, not antagonistic, attitude as the private sector may be more attuned with the market’s expectations. Both the private and public sectors should jointly deliberate and coordinate their programs in monitoring the movements of the economy, as well as the resolution of problems.
****
Leonardo A. Lanzona Jr. is professor of economics at the Ateneo de Manila University and a senior fellow of Eagle Watch, the school’s macroeconomic research and forecasting unit.