The Growing Woe

Falling leaves usher change,
No good news in range,
For the external sector.
Slow growth is the factor,
Affecting key items of export-import.
This is before flu outbreak report.
As the pandemic turns worst,
They are still calculating the cost.

Gradually, damages caused by recent flu pandemic are surfacing. But they are still in initial stage and far from sketchy details. Maybe a week later a comprehensive assessment will be available. Meanwhile, I tried to look into the external sector and tried to find out what the key contributors to import and export baskets will look like in the next few years. First I tried to look at letter of credit opening and settled as reported by Bangladesh Bank. I found that data on first, second, third and fourth week of January were reported. Even the second week of February was reported. However, to my dismay, I could not find data on fifth week of January and first week of February on Bangladesh Bank website. It was when coronavirus was making headlines. So my endeavour to see coronavirus' initial impact on import and export turned into a damp squib.m

Then I turned towards the export and import data between 2013 and 2019 as reported by Bangladesh Economic Review 2019. I was to see what the key commodities' import and export figure would look like couple of years down the line provided the information in the aforementioned period. For export, I picked up woven /readymade garments, knit , frozen food, agri and engineering products. For import, I had chosen rice, wheat, cotton, yarn and capital machinery. BER 2019 reported the data for 2019 till February /March 2019 only. So all the data for 2019 reflected data available till February /March 2019.

Source: Bangladesh Economic Review 2019

Then I took natural log of these products. This gave the percentage change of these items over the years. Then I took the first difference of the natural log variables. This gave the idea how the percentage change would act overtime. Moreover, the first difference would make the variable stationary, it was assumed. However no tests for stationarity and autocorrelation were carried out.
Then I ran a VAR model on lagged values of these variables. For instance, for export the VAR(1) would look like this:
🔺 lnExp_t=a + b🔺 lnExp_t-1

For other variables pretty much similar regression was constructed. Having run the model, import, export, cotton, knit, woven and engineering variables showed that their model parameters were significant, based on chi² readings. For some variables 1-lag was considered. For others, both 1-lag and 2-lag were considered.

Then a 4-step ahead forecast was carried out. It appeared that percentage change of import, export, knit, woven, engineering would decline in future. It would take a plunge for export, knit, woven and engineering. But these forecasts need to be taken with a pinch of salt as initial diagnostic check was not done and other variables were not included in the individual regression function. Nevertheless, the forecasts gave a visual presentation how percentage change of key contributors to external sector would behave in future based on their past values.

Bangladesh already witnessed a - $9.7 billion current account deficit in 2017-18. It sustained in subsequent years. So things are already in bad shape for the external sector before recent flu outbreak. Key import and export items registered a slow growth. Coronavirus outbreak will evidently make a dent in the external sector. How worst the damage could become is yet to be assessed.