no pat answer, no grapevine

一見正しそうなことや噂になんか流されない。

Can statistics tell the truth? --some tips and techniques for interpreting statistics--

    To grasp what is happening now in the economy is the ever-lasting challenge a lot of people have engaged in. One of the most popular tools is “economic statistics” like GDP, unemployment rate and so on.  Such statistics tell us how rich we are or how many people are seeking their job at particular time. Therefore, not only economists but also ordinary consumers pay attention to results of economic statistics released by statistics bureau worldwide. On the contrary, some people hesitate for looking at statistics, because there are only numerous numbers…

    Economic statistics are not always telling the truth. It doesn’t mean that you should not trust statistics itself, but that recognize their limitations. Though a lot of statisticians have tried to maximize the accuracy and minimize the error, it’s on a journey. Why? There are at least two possible reasons. First, under budget constraint, it’s difficult to have all the people or companies answer questions. Second, it’s extremely difficult to ask questions without causing misunderstandings.

    You may say “it’s no use judging or deciding policies based on statistics“, but they certainly show the reality of our life in a particular aspect to some extent. Or else, suppose we get over statistical limitations. Now, you can no longer avoid understanding what statistics are saying. Following, let’s think about how to interpret statistics together.

Figure 1. is the development of nominal GDP (original series) in Japan. The latest quarter (2012/10-12) value is 123,170.7 billion yen, while the last quarter (2012/7-9) is 115,888.5 billion yen. The percent change from previous quarter is about 6.28%. Can you believe this? Is it really that Japan’s economy started to recover?

fig.1 : the development of nominal GDP (original series) in Japan


    As you see, the graph shows regular cycle, the value jumps up in October to December quarter every year. This is because “private consumption”, one of the most important GDP components, tends to increase its value in October to December quarter (see. fig.2). It implies that just because the figure in 2012/10-12 is larger than that in 2012/7-9, the economic condition in Japan doesn’t necessarily get better.

fig.2 : the development of private consumption (nominal, original series) in Japan