First of two parts
IN the wake of the lower-than-expected economic growth in the third quarter of 2014, the contribution of government spending to the economy has become the focus for analysts and policymakers. Retracting 2.9 percent year-on-year in the third quarter, government spending, or rather the lack of it, was blamed for slowing growth, a problem that seems to have continued in the fourth quarter.
As has been pointed out in several reports, the slowdown in GDP growth in the latter half of 2014 certainly appears to have been much broader than can attributed to shrinking government expenditures, but government spending does have an impact on the economy. “Government spending is about 10 percent of the entire economy’s output,” Dr. Jose Ramon G. Albert, senior research fellow at the Philippine Institute for Development Studies (PIDS) pointed out, “So increased government spending will certainly have an effect.”
Dr. Albert, who is probably one of the country’s foremost experts in handling socioeconomic statistics, suggested that the real impact of increasing government spending could only be reliably estimated after a deeper analysis of economic inputs and outputs. The implied uncertainty, however, has not stopped some analysts from making optimistic estimates of the growth dividend of “improved” spending.
In a report released just before the end of the year, Citi Research estimated that increased spending for the Typhoon Yolanda reconstruction program would increase economic output by 2.2 percent over the next three years. In a mid-year report, the Center for Business and Economic Research at the University of Nevada at Las Vegas took an even rosier view, pegging 2015 growth at 8.4 percent. Moody’s Analytics, on the other hand, took a glass-half-empty perspective in its year end assessment of the Philippine economy; having earlier forecast a moderate 6.5 percent annual growth rate in 2015, in its latest credit outlook the agency warned “the government’s real GDP growth target of 7 percent to 8 percent for 2015 will be difficult to achieve if budget release and use are not improved.”
The general impression is that the Aquino Administration’s handling of the budget, which is assumed must improve in 2015, will add something between half a percent and 2 percent to the GDP growth rate this year due to increased spending. A cursory analysis of the available data, however, suggests that not only is a significant increase in spending unlikely, any increase is likely to have far less of an impact on the overall economy than anticipated.
Inconsistent historical performance
Since 1999, the proportion of GDP attributable to government spending has fallen within a relatively narrow range, from a high of just over 12 percent to a low of 9.3 percent. The long-term average over the past 15 years has been 10.26 percent; spending by the Aquino Administration has been just slightly higher than that at 10.3 percent. The biggest increase in spending so far during the second Aquino era, a 15-percent acceleration which occurred in 2012, budged the proportion of government consumption to total GDP by less than one percent, from 9.85 percent in 2011 to 10.65 percent in 2012 (chart 1).
The year-to-year change in government spending has been erratic although it has generally trended upward (chart 2). Changes in spending, however, do not always correspond with a similar movement in GDP. In 2003 for example, the Arroyo Administration increased the growth of spending by nearly 7.5 percent, and the GDP growth rate accelerated as well, from 3.65 percent in 2002 to nearly 5 percent in 2003. The next year, by contrast, the rate of spending growth slowed again, but GDP growth soared to 6.7 percent.
The same somewhat contradictory pattern was repeated between 2005 and 2007, and in 2009, the year the most serious effects of the global financial crisis were felt here, a sharp increase in spending failed to stimulate the economy; the economy still grew, but only at 1.15 percent, three percent slower than in 2008. Likewise, in 2013 the growth of government spending slowed sharply from 15.5 percent to 7.7 percent as political scandals erupted, but the overall economy moderately increased its growth pace from 6.8 percent to 7.2 percent.
How government spending affects the entire economy can be simulated by analyzing the economic input-output tables, according to PIDS’ Albert; that was the methodology used by Citi Research in developing its recent estimate of the expected impact of Yolanda reconstruction spending.
Analyzing an input-output table – which is a form of “what-if” analysis – is relatively straightforward. The problem is that the most recent data available is nearly ten years old; the last time an input-output table was created was in 2006. That is because collecting the data to build an input-output table is a herculean task; the ‘current’ I-O table for the Philippine economy comprises 240 separate economic sectors, for each of which accurate data on consumption and production must be gathered.
Appreciation of the difficulties in producing an I-O table notwithstanding, any forecasts based on ‘dated’ data will have some degree of uncertainty. The economy has undoubtedly changed since 2006, but to determine in what ways and by how much would require a new I-O table; otherwise, the best any analysis can provide is an educated guess. For example, Citi Research based its calculations on a multiplier of 1.65; in other words, each peso of government spending would result in P1.65 of total economic output. According to the 2006 I-O table, the multiplier of government spending across the entire economy at that time was 1.56, of which 1.09 was secondary government spending (meaning that every peso of spending in one part of the government requires P1.09 in output from the rest of the government).
That suggests that a realistic multiplier might be closer to 0.49 – the difference between “government” and “rest of the economy” inputs to meet government spending demand, plus a small margin of about five percent to account for growth since 2006. In that case, the P57 billion in extra reconstruction spending each year that Citi Research forecast to result in P94.3 billion in additional income would instead produce a more modest P27.9 billion, adding just 0.22 percent to GDP growth annually, or about 0.66 percent over the three-year window studied by Citi.
Just as the assumption that government spending boosts the overall economy seems to be based on less than compelling evidence, the opposing view is likewise very debatable, as we will discover in part two of this special report on Tuesday.