Growth forecasts in 2012-13 Budget
Mr Muhith brought down his fourth budget last week. For his first three budgets, I published a number of articles with my erstwhile Drishtipat colleagues (note to self: set up an archive of published analysis). These days, I prefer to focus on the blog. So instead of writing op eds for the Daily Star or BDnews24, I’ll do a number of posts over the next few weeks. Dear reader, this means I am writing exclusively for you.
The typical analysis of the Budget one sees in the media does not look at the past performance. And yet, analysis of how the government’s previous Budget forecasts turned out is a good way to judge the current forecasts. Using a set of charts, this post focusses on the government’s successive forecasts of GDP growth.
The first chart shows GDP growth in the past two decades, and Budget forecasts out to 2016. The dotted red line represents the outcomes. When the government came into office, the economy had been slowing for a couple of years — first because of political uncertainty, and then reflecting the global recession. Mr Muhith’s first budget – in June 2009 for FY10 (that is, 2009-10), represented by the solid green line – projected a further slowdown before a rebound.
In the event, FY10 growth was stronger than expected, and the government’s second budget — the green dotted line — projected continuous acceleration in the economy. At 6.7%, economic growth in FY11 came in with what was projected in the Budget. Nonetheless, the third budget — fluorescent green dotted line — projected a more moderate acceleration in FY12.
Whereas in June 2010, the government was forecasting the economy to grow by 7.2% in FY 12, a year later the forecast was revised to 7%. In the event, the BBS estimates the economy to have grown by 6.3% in the year.
This modest soft patch notwithstanding, the latest budget — solid red line — forecasts the economy to rebound and follow the same trajectory as in last year’s budget. Growth is expected to be 7.2% in FY13 – if it comes to pass, this will represent the fastest acceleration in over a decade.
Is this too optimistic?
I’ll tackle the specific economics behind the forecast later. Here, let me note that on average, budget forecasts have had a rather small optimistic bias over the past decade — in technical terms, budget forecasts over the past decade have had a mean forecast error of 0.02 percentage points.
Graphically, this is shown in the chart to the left. Here, the horizontal axis shows budget forecasts, and the vertical axis shows the actual outcome. Green dots represent years. Dots on the (grey dotted) 45 degree line means that in that year, the budget forecasts were accurate. Dots above the 45 degree line means the actuals were stronger than forecast — that is, forecasts were too pessimistic: this was the case inFY10, for example. Similarly, years where the budget forecast was too optimistic are below the dotted line.
The chart shows that under the three different governments, there is no particular bias in terms of over-optimistic forecasts. There has been years when the economy performed better than the budget forecasts, and in other years things have not gone as well. Overall, it’s broadly a wash.
Of course, this doesn’t mean the forecasts are accurate. They plainly are not usually — althouth FY03, FY05 and FY11 were mighty close.
Turns out that over the past decade, the government’s forecasts have been more accurate on average than the ADB forecasts (I haven’t had the time to crunch the numbers for WB or IMF, but ADB’s analysis is usually more detailed, so I doubt the other two will be more accurate). In technical terms, the mean absolute forecast error for the budget forecasts has been 0.38 percentage points, against the ADB’s 0.45 percentage points, over the past decade.
This chart compares the budget and ADB forecasts against history. Actuals are represented by the yellow columns. The red squares are the budget forecasts, with error margins around them. The green triangles are the ADB forecasts, with their error margins. Up until the FY06, there were little between the government’s and ADB’s forecasts, and the actuals. For FY07, the government forecast was much more optimistic than the ADB’s. In the event, FY07 growth came in halfway between the two forecasts.
For the two following years, the two forecasts were virtually same, and both proved too optimistic. In more recent times, the government has been ever more optimistic than the ADB, and in FY12 both have proved to be too optimistic.
For FY13, while the government expects a very sharp acceleration, the ADB expects growth to moderate even more, to 6%. This is the farthest two forecasts have been in the past decade.
Is the government too optimistic, or the ADB too pessimistic?