Macroeconomic Policy: AS and AD Curves

If an economy has a flat AS curve, implying that AD policy would be more effective to stimulate growth, it remains unclear what type of AD shock should be pursued. When the resulting output growth and prices are linked with the poverty line and incomes of the poor, the uncertainty in terms of policy implications on poverty gets even bigger, e.g., raising government expenditure will generate different outcomes of growth-poverty nexus than lowering the interest rate.²

Two irregularities prompt investigation of the AS and AD curves' slopes: first, a standard ADbased policy is not always effective when used to counter a major shock like the Asian financial crisis that occurred in 1997; second, AD management policy to control inflation may not work effectively, making a combination of slow growth and high inflation possible, both of which tend to worsen poverty conditions. This partly explains why the predominantly ADbased policies of the financial crisis have led to a significant increase in poverty. The outcome of the policy response in Thailand (where the Asian financial crisis originated) and in Indonesia (where the effect of financial crisis was most severe) has not been the same; less effective in the latter than in the former. The post-crisis experiences of the two countries also differ: Thailand has achieved a non-inflationary high growth pattern while, as of 2007, Indonesia is still trapped in a slow-growth high-inflation mode.

Central to the diverging outcomes of policy response are the characteristics of the outputprice relations in the two countries. Under normal circumstances, a line stretching from northwest to southeast quadrant is generated under AS shocks, and a line from northeast to southwest quadrant is generated under AD shocks. However, simply plotting the data of quarterly growth rates of real gross domestic product (GDP) and inflation will not provide information as to whether the locus shifts are due to AD or AS shocks. To the extent that both shocks jointly determine the changes in output and prices, a decomposition procedure needed to be applied. This could be done either through a univariate approach of autoregressive integrated moving average (ARIMA) with the assumption that disturbances are either orthogonal (Watson, 1986) or serially correlated (Beveridge and Nelson, 1981), or, through a multivariate approach. This study used the latter by adopting the structural vector autoregression (SVAR) with the Blanchard & Quah (1989) (hereafter B-Q) restriction and a decomposition technique used in Gamber (1996). Appendix 1 [ PDF 53.4KB | 2 page ] outlines the model and the procedure in more detail.

I used the 1993:Q1–2007:Q2 data of real GDP and consumer price index (CPI) (with 2000 as the base year) for Thailand and Indonesia to measure the slopes of AS and AD curves in the two countries. Unit root test using (Augmented Dickey-Fuller [ADF]) suggested that when the second-difference log is used, the null hypothesis of unit-root is rejected at 1% level (Appendix 2 [ PDF 40.7KB | 1 page ]). To determine the appropriate time lag, the Ljung-Box test was used for the selection process, from which it was found that all residual series will no longer be correlated when a lag length of 4 is used (Appendix 3 [ PDF 44.7KB | 1 page ]). Thus:

The results show that, in both countries, the slopes of AS and AD are according to what the theory predicts, i.e., positive for AS and negative for AD (Figure 1 [ PDF 416.1KB | 1 page ] and Figure 2 [ PDF 416.1KB | 1 page ]). The slopes in both countries are clearly flat—even more so when compared to the slopes in Republic of Korea (hereafter Korea) and Malaysia.³ This suggests that in Indonesia and Thailand a positive AD shock would have been effective to stimulate non-inflationary growth during the period of observation; output would have increased faster than prices. Along the AS curve, a positive growth innovation of 1% corresponds to a positive inflation innovation of 0.01% and 0.02% in Thailand and Indonesia, respectively.

The decomposition results also show that Indonesia has a steeper AD curve (with a slope equal to -1.474). Along the AD curve a positive inflation innovation of 1% corresponds to a negative output growth innovation of -0.68%. The corresponding figure for Thailand is -1.25. To the extent that a stabilization policy tends to focus on inflation control, a steep AD curve suggests that using AS shock rather than AD policy to lower price levels would have been more effective.

Broken down into pre- and post-crisis, Table 1 [ PDF 44.1KB | 1 page ] shows that the slope of the AS curve in both countries became much flatter after the crisis, i.e., from 0.590 to 0.008, and from 0.316 to - 0.133 in Thailand and Indonesia, respectively.⁴ In the case of AD curve, however, the trend was the opposite: it became flatter in Thailand (from -0.919 to -0.686) and steeper in Indonesia (-0.819 to -1.008). This clearly suggests that after the crisis, an AD-based policy would have been even more effective than before the crisis in stimulating growth and less effective for controlling inflation. This finding is most profound in Indonesia because not only has the slope of the AS curve turned negative, but the AD curve has also become much steeper. In Thailand's case, the post-crisis AS curve slope remains positive, and the negative-slope AD curve has become less steep.

Capacity utilization during and after the crisis is usually low (i.e., the gap between potential and realized output is large), in which case an expansionary policy is needed. As shown in Figure 3A [ PDF 44.1KB | 1 page ] and Figure 3B [ PDF 101KB | 1 page ], this was indeed the case in Thailand and Indonesia, respectively.

Comparing the findings in Figures 3A and 3B with the actual policy in both countries validates the predicted outcome. As Figure 4 [ PDF 101KB | 1 page ] and Figure 5 [ PDF 53.6KB | 1 page ] show, immediately after the crisis Thailand adopted an expansionary AD policy, the fiscal deficit widened, and the interest rates were lowered. The expansionary fiscal policy continued even when the external assistance provided under the Miyazawa Fund and other sources ended in 2000. In 2001 and 2002, the fiscal deficit was recorded at between 2% and 3% of GDP. As the economy recovered, fiscal surpluses began to appear in 2003.

By contrast, Indonesia's fiscal position was in surplus even in the early stages of the crisis and its tight monetary policy lasted longer despite the downward pressure on the output during the time. Only a few years later, a fiscal deficit began to appear. Until 2002, as a percentage of GDP, Indonesia's deficit was not only lower than Thailand's but also lowest among all the Asian countries hit by the crisis. This is quite puzzling given that Indonesia suffered the most from the shock, with the sharpest fall in output.⁵ While the interest rates in Thailand were lowered in 1997, the rates in Indonesia were raised to 17% in 1997 and 37% in 1998. Since then, the rates have continued to be at double-digit levels, except in 2003 and 2004, when Indonesia's rates were a lot higher than Thailand's (Figure 6 [ PDF 53.6KB | 1 page ]).

Thus, while Thailand's policy was consistent with what the decomposition analysis suggests, Indonesia's policy was not. The resulting outcomes were as expected: Thailand's economy recovered more steadily than did Indonesia's.

Following the decline during 1996:Q4 and 1997:Q1, Thailand's real GDP rebounded briskly in 1997:Q2, peaked in mid-1997, then fell more than 10% before reaching a trough during the second half of 1998. The inflation rate also surged, reaching 9% in the first quarter of 1998, then a double-digit rate in the second quarter, before declining to 5% in the last quarter (Figure 6 [ PDF 53.6KB | 1 page ]). As expected, the poverty line and incomes of the poor (hence the poverty incidence) was adversely affected (see Figures 25 and 26 in Section IV).

Indonesia's GDP growth, on the other hand, has been the most disappointing among all Asian crisis countries (Figure 7 [ PDF 71KB | 1 page ]); its largest GDP fall was in 1998, but Indonesia's turn around has been the slowest. The government's decision to inject a huge amount of liquidity support to some troubled banks led money supply to increase significantly, despite the high interest rates. As a result, investment fell and prices soared. The IMF-recommended policy of structural change (AS policy shock) failed to produce the necessary recovery because the AD was severely curtailed.

With real GDP falling by 15% between the third quarters of 1997 and 1998, inflation surged, the unemployment rate increased persistently (Figure 8 [ PDF 71KB | 1 page ]), creating a double whammy: lower incomes of the poor and rising prices and poverty line. This was why Indonesia's poverty incidence soared dramatically during that period (Figures 25 and 26).

Indonesia's inflation rate has been the highest among the crisis countries. This is despite its relatively tight monetary policy (Figure 6 [ PDF 53.6KB | 1 page ] and Figure 9 [ PDF 112.4KB | 1 page ]). The AD-based policy has clearly been less effective. As indicated earlier, the country's AS curve was flat and the AD curve was very steep. Yet, all indications point to a strong tendency for the authority to continue using the AD-based policy to curb inflation at the costs of growth, income, and poverty. Indeed, during the crisis the negative impacts of the policy responses on poverty has been much more severe in Indonesia than in Thailand.

The role of the supply and demand shocks as the source of inflationary pressure can also be analyzed by generating the time series of inflation due to each shock. The results are shown in Figure 10 [ PDF 112.4KB | 1 page ] and Figure 11 [ PDF 124.1KB | 1 page ] (excluding the drift term that represents the persistent impacts of the supply shock). The reconstructed time series components clearly show that in both economies the supply shock dominated the source of the sharp fluctuations of prices in 1997. In Thailand, the domination occurred from 1997:Q2 to 1998:Q3, while in Indonesia it lasted longer, i.e., from 1998:Q2 to 2002:Q4. By far, Indonesia's price increase was the most dramatic as the country's socio-political crisis and major institutional changes prompted a major cost-push pressure. The severe drought season related to the El Nino weather phenomenon also exacerbated inflationary pressure during that time. At any rate, controlling AD to curb inflation in such circumstances was clearly ineffective.

Download this Discussion Paper [ PDF 978KB| 34 pages ].

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [1] comment(s) for this entry. Post a comment.

1. Ruly
   (posted 11 July 2008 / 02:08:15 PM)
   
   This paper gives me a new insight about the real condition of my country economy, and how the authority (government) had applied many conventional economic policies, which is not suitable for Indonesian economy that has different characters and conditions from the west, the origin of these conventional policies..
   
   Furthermore, I was very lucky because the author presented this paper in my class as a guess lecturer about 2 months ago. For the first time after 1.5 years studying economics, a lot of my "unanswered" questions being answered, and I even didn't have to ask..