Non Sequitur Economics
In this essay from Singapore many years back, I examine the non sequitur economics specifically on the consumer price index (CPI) and inflation rates.
Chua Mui Hoong’s recent article in the Straits Times (Review, 24 Nov 2006) contains a spirited defence of the various fee hikes instituted by the government since this past general election. The article considers the history of fee hikes after each general election, and attempts to refute the politically “seductive argument” put forward by “cynics” to the effect that fee hikes have always been the consequence of PAP win at the polls.
There are three types of lies – lies, damn lies, and statistics.”
— Mark Twain
Her basic argument is as follows:
“If price hikes were severe after elections, then inflation rate should go up a year after elections. But the data show otherwise. Since 1980, the inflation rate has consistently gone down one year after all elections. The exception was the 1988 elections. That year, inflation was 1.5 per cent. A year later, it went up, to 2.4 per cent. But after the 1980, 1984, 1991, 1997 and 2001 elections, inflation rate fell the following year by 0.3, 2.1, 1.1, 2.3 and 1.4 percentage points respectively. In fact, after the 1998 and 2001 elections, the inflation rate was negative, suggesting a fall in prices across the board, as the Asian financial crisis and the spectre of terrorism kicked in. While crude inflation rate data may not give a finessed enough picture of the extent of government fee increases, it does show quite conclusively that the overall price increases was hardly crippling, but moderate or negative after elections. So inflation history vindicates the Government from the charge that it saddles the people with fare hikes after each election.“
The question is: just how sound is this reasoning. To get a handle on Chua’s argument, we need to take a closer look at her use of economic indicators such as the Consumer Price Index (CPI) and Inflation Rates.
CPI Index and Inflation Rates
The Consumer Price Index (CPI) is a popular measure of price inflation in retail goods and services and the market sensitivity of this index is very high. In simple terms, the index measures the changes of prices in a fixed basket of goods and services purchased commonly by the majority of households. The goods and services measured include food, clothing & footwear, housing, transport and communication, education and stationery, healthcare and recreation & others. Since 2004, the Singapore Department of Statistics (SingSTAT) has adopted 5170 brands and varieties for the computation of the CPI (see  and ). The coverage of the statistic takes into account 5400 households which constitute the middle 90% of the income distribution with the remaining tail ends truncated. The rationale for that truncation is the consumption habits for the very poor and very rich differ vastly from the middle 90%.
Since the CPI measures the changes in prices of goods, the index can be adapted to calculate the inflation rates of the nation. It is important to understand how this value is calculated. The monthly change in the CPI is calculated based on the difference in the index of the present month and the previous month that provides an useful indicator to measure short term price changes. However, to measure the annual changes, the index of the month in the present year (say, August 2006) is compared with the index of the month in the preceding year (August 2005). The annual inflation rate is inferred by comparing the average for the 12 monthly indices with that of the preceding year. To give us some perspective, we obtained the CPI index and inflation rates  and plotted them from 1961 to 2005. It is immediately obvious from figure 1 that the CPI increases from 1961 to 2005.
As Chua has rightly indicated, if you look at the fee hikes imposed by the government after each election cycle, there is a drop in the inflation rates in the year after each of the elections in 1980, 1984, 1991, 1997 and 2001 (see Figure 2). In fact, to add to what she says, it also turns out that inflation fell from 3.1 to 1.7 in 1994, the year after GST was first instituted, and from 1.7 to 0.5 in 2004, a year after it was raised to 5% (on the other hand, it increased from 0.5 to 1.7 in 2003, a year after GST was raised from 3 to 4%).
It is interesting why this should be happening. In her view, the common man’s perception of price increases is different from what reality really is.
But one should be careful when interpreting reality using economic indicators. A naive inference from increasing the costs of living will mean that the inflation rate will also go up. So, why is the inflation rate dropping even if the costs are increasing?
An answer may be found in market efficiency. As an illustration, consider the electronics gadgets industry. In this sector, prices are always decreasing when better models emerge. For example, a 4 Megapixel digital camera cost $1K two years ago. Today, you can get one with 8 Megapixel for $500. Depending on the model of the television, computer or digital camera, it’s not difficult to think that the overall prices of these products will drop. This also means that the Singaporean market in such items is efficient in coping with competition. On the down side, market efficiency could also be the very reason why small medium enterprises (SMEs) are not coping well against the big multi-national corporations as the latter are more able to control price changes, making it difficult for the SMEs to compete.
Unemployment Rates and Inflation
One funny thing about Chua’s argument is this: it’s actually easy to construct an analogous argument to justify the claim that the economy is doing well on the basis of price hikes. Now, it is well known in economics that the unemployment rate is correlated with the inflation rate.
If you examine the sudden surges in inflation rates, you will see that these were also the years which Singapore was in the economic boom cycle (look at the periods between 1973-1975, 1980-1983 and 1990-1997). So, wouldn’t the low inflation rate plus the historically higher unemployment rate in the past few years show that the economy is not doing well?
If the opposition party argues on that we should have low inflation rates, should then we tell them that such an outcome will mean a higher unemployment rate. Conversely, for Ms Chua, a low inflation rate does not really mean the economy is doing well.
Connecting the dots: Is Inflation really bad for Singapore?
An intuitive question to ask is how inflation  gets started in the first place, and whether it is really detrimental to the growth of the economy. It depends on who you believe. If you are in the monetarist camp, excessive growth in the money supply is responsible for sharply rising prices. If the supply of money is increasing faster than the output of goods and services, it just means that there is just too much money for too few goods and services. Since there is a relative scarcity of goods and services relative to money, prices will rise. To complete the picture, the reason for excessive growth in money supply is also attributed to high employment rates. In the good times, companies hire more people and hence there is more money flowing around. It is not puzzling why inflation and employment are correlated this way.
If you subscribe to a Keynesian view–which argues that when the overall demand (from customers, businesses, government and foreign investors) greatly exceeds the economy’s ability to meet its requirements, the resulting shortage in supply will accelerate the price of goods and drives inflation at a higher rate–the imperative is to determine how fast inflation will go up. That will depend on where the economy is positioned in the business cycle. Perhaps somewhat counter-intuitively, if there is a sudden surge in demand for goods and services following a recession, it does not automatically imply that inflation has taken place. Rather it is the economy performing a self-correction towards achieving market efficiency. Inflationary pressures can be felt in the later cycle when the scarcity principles come knocking on the door once again.
Is inflation really a bad thing? The problem with inflation is that it generates a climate of instability and distorts people’s view of the economy. The dot com bust in 2001/02 is an example of this situation: the IT companies were overly hyped by the investment banks with the IT workers demanding a higher pay to offset increasing costs of living. From another perspective, the higher prices offer companies the opportunity to generate more profit which in turn gives the workers better bonuses. Better still, the governments can exploit the chance to generate more tax revenues (particularly thru the GST route). People with the strong tendency to borrow more can pay their loans back with cheaper currency.
But it is hard to tell whether it is good or bad for the economy. The moral of the story, as we present to you, is to be skeptical of people from both sides of the house in conjuring non sequitur economics to justify their perspective of the Singaporean economy.
 The figures for the plots for the Consumer Price Index and Inflation Rates are obtained from the Singapore Department of Statistics (SingStat). About the details in calculating and interpreting the CPI, please look at this URL and document from SingStat. The unemployment figures (from 1993 to 2005) are extracted from the Ministry of Manpower (MOM) website.
 Bernard Baumohl “The Secrets of Economic Indicators: Hidden Clues to Future Economic Trends and Investment Opportunities”, Wharton School Publishing.
 See Phillips Curve in Wikipedia.