Predicting economic turning points

They say that death and taxes are the two unavoidable things in life. To that, we should add economic crises. Being able to predict when the next downturn will be is worth a lot of money, literally. For a government, it could be a matter of survival.

I DO NOT KNOW about you, but I would like the economy to be just good; stable and rising steadily. But in reality we know that this is not the case.

We know all about the business cycle, when economic expansion reaches a peak and then slumps into a recession. From the trough, the economy starts to grow and we do it all again. While we’re aware of peaks and troughs, do we know when these are going to occur? Bad times seem to sneak up on us really slyly and mess up all the big plans we have. If only we know in advance, we could make necessary adjustments to prepare for the changes. If we have this intuitive “power”, we could then go on the “counter attack” and face the economic situation with appropriate policies. Society would then be less affected by economic instability.

Personally, I am more concerned with my wallet especially if I am to be able to afford the new iPhone 5G that is rumoured to be released this year. Regardless, we should all be paying attention to the economy, it affects us all. From the Pak Cik in the market who sells nasi lemak to the mortgages we’re paying off , what we do affects the economy and what happens to the economy affects us too. If the economy is in a recession, it will determine if we have rice or porridge for our next meal.

Economists have long been trying to detect when the economy will reach its next turning point. Salih Neftci’s Probability Method is one constructed to calculate probabilities of approaching cyclical swings. His method seems superior in the sense that it is able to screen off false alarms (so we don’t have to eat porridge when we can eat rice!).

This article was written following an evaluation undertaken in Penang to determine the signal of the turning point and to observe its occurrence relative to the “actual turn” (by referring to the pressure curves 3/12 and 12/12 in Figures 2 and 3).

Some may argue that if it works in many other countries (the US, UK, Germany and Japan), it should work in Penang too. However, we have to be aware that each nation is unique. Further studies have shown that even leading indicators combined with Neftci probabilities that worked best for the US did not work as well for the UK. For Penang, the Industrial Production Index (IPI) is chosen as the leading indicator to observe economic performance. Also, compared to other data like the gross domestic product (GDP) which is revised for the whole year, the IPI is easily available through the Department of Statistics and is relatively up to date, with only a lag of two months.

The Neftci Probability Method is based on sequential analysis and draws upon three pieces of information. This information can be obtained by studying past data. First is the probability of a recovery in the economy; second is the probability of a recession; and lastly, the combination of the two probabilities with the last month’s probability estimate. The formula is:


The IPI is tested with the method, and from the prediction in Figure 1, we are in the phase of downturn in the economy and in January 2011, we should be expecting a turning point as we enter the recovery phase. The data obtained from the Department of Statistics confirms the prediction. In Figure 2, we see that the economy is declining from the peak, and it is moving towards the trough of the economy.

This simple observation is only a preliminary exploration into how the method can be used to Penang’s advantage. What is important from this exercise is that we are able to pick any “normal”1 point from the graph of the leading indicator, IPI (Figure 3), and detect its position in the business cycle, whether it is at the expansion, recession, contraction or revival stage. The threshold of 110 in Figure 2 marks the 50% probability in the sequential analysis, and if the threshold of 110 is prior to a peak in the economic performance, the probability is that the economy is in a recovery.

It might not be exciting to go over graphs and the method, but for me the findings give me the justification (and eases my guilt!) for booking the new iPhone and perhaps, even the iPad that I have been eyeing for a while. Numbers are just figures. But put together, they form patterns which tell stories. If we pay attention to these, the data forms a window to the future which can serve as a guide in our endeavours.

There is a catch though—numbers have to be ready for policymakers and investors ahead of time so that they can react to the calculations.
1 Normal” refers to economic conditions where there is no exogenous factor aff ecting the economy.

Pauline Khoo is a Penangite currently studying in Hollins University, Virginia. She is pursuing her Bachelor of Science in Mathematics and Economics. She hopes that she can contribute to the study of economics in the near future.

Related Articles

Jun 2010

Celebration of culture

The George Town World Heritage Office is set to win the hearts and minds of tourists and Penangites.

Sep 2015

PARK(ing) Day: Reclaiming urban space, one bay at a time

Penangites get the chance to be creative with their public space.

Nov 2016

Malaysia is Like a Plate of Nasi Kandar

From its origins as a simple meal to fill the tummy, nasi kandar has become a Malaysian institution.

May 2017

Hin Bus Depot – Derelict No More

From ruins to creative hub, the art cluster finds its footing.