Nowcasting Bangladesh's economy
The quarterly national accounts can help assess the economy in real time
There has been a lot of attention on the exchange rate and the Bangladesh Bank’s stock of foreign reserves. Every other day there is a media report or youtube video on the true value of the reserve or where the exchange rate might end up. This is understandable. Less explicable is the relative ambivalence about the Bangladesh Bureau of Statistics' newly released quarterly national accounts. This new series was one of the conditions of the IMF program announced last January. It is a condition the authorities have met.
The quarterly figures deserve to be analysed in far greater detail than has been the case thus far. There are many good reasons to doubt the accuracy and credibility of BBS estimates. The quarterly national accounts should improve matters. For one thing, quarterly accounts are mechanically much harder to fudge than the annual ones. Further, we can cross-check the sectoral contributions to the overall growth against other indicators. And thirdly, analysing these quarterly figures can potentially improve our understanding of the economy.
Let’s look at the figures. Chart 1 shows year-ended growth in GDP(P) over 28 quarters between Sep 2016 and June 2023. A few things immediately stand out that are less obvious in the annual data.
Firstly, according to the annual figures, the economy continued to grow throughout the pandemic. This always seemed counterintuitive (if one is to be generous and not at all cynical about the government manipulation of figures). The quarterly data show the pandemic’s impact on the economy much more clearly. The economy grew by 8 percent in the year to March quarter 2020. In the year to the next quarter, it shrank by 7.7 percent. There were hardly any growth in the rest of 2020. This is much more in line with people’s lived experience of the lockdowns and reopenings after vaccines became available.
Secondly, the recent economic woes are also very clearly visible in the quarterly figures. The economy slowed sharply in the June quarter 2022 — that’s when severe import controls were put in place, and it is clearly the industry sector that suffered most heavily. This again accords well with anecdotal evidence. The economy, however, grew strongly in the remainder of 2022. Recall, this is when the depreciating taka would have helped households receiving remittances (even if they were sent through informal hundi channels), while caps on borrowing rates meant the real interest rate was negative for people who could access credit. This probably explains the strength in the services sector in the second half of 2022.
In the first half of 2023, however, services took a battering. This is when inflation really bit into household budgets, and they responded by cutting non-essential expenditure. Services hardly grew in the year ending March quarter 2023, with tourism, trade and entertainment declining — we can see this is in Chart 2.
Industry sector, however, held up in fiscal year 2022-23. This is not surprising. Manufacturing grew because of exports, which remained steady. And construction probably benefited from all the shiny stuff the government had been building.
One would miss all of these nuances by looking at the ‘annual GDP growth of 6 percent’. However, as shown in Chart 3, even annual (fiscal year) growth in the GDP(E) can reveal a story that is consistent with the above.
Domestic private demand — household consumption and business investment — slowed significantly in 2022-23 fiscal year. This is in line with the trends in the services sector seen above. Business investment contributed only 0.4 of a percentage point to growth last year, compared with an average of around 2 percentage points before the pandemic. The weakness in business investment is very much consistent with the anecdotal evidence and media reports. In contrast, net exports made a positive contribution to growth last year — in line with the strength in industry sector and the severe curtailment of imports that are visible in the trade statistics.
We can also a do a simple thought experiment to see the effects of the recent shocks on the economy. Before the pandemic, real GDP used to grow by about 2 percent a quarter. Suppose that trend had continued from March quarter 2020. According to the BBS, the economy had recovered to that hypothetical trend by June quarter 2021, but the recent shocks had opened up another gap. This is shown in Chart 4.
We can, in fact, go even further. We can do two things. One, we can further explore the veracity of the sectoral growth figures, and their implications for the expenditure side, by looking at partial data such as sales or production figures. For example, perhaps the weakness in the March quarter 2023 was sharper, or more prolonged, than what the BBS might be reporting. This is an exercise I will leave to people who are closer to the ground in Bangladesh.
The other thing we can do is to use the quarterly growth figures, together with a few other indicators that are readily available with higher frequency than the national accounts, to guesstimate the pace of economic growth in real time. Chart 5 compares the year-ended GDP growth with year-ended growth in electricity sales, industrial production, credit to private sector, and taxes.
Intuitively, each of these variable should be correlated with GDP growth. And we do see that overall, they do tend to move somewhat close together — but the mileage varies considerably. Chart 6 goes a bit further by comparing year ended growth in GDP (y-axis) with year-ended growth in each of these variables (x-axis) bilaterally.
This simple exercise suggests that if we knew recent trends in industrial production, and to a lesser extent tax collections and electricity, we might be able to guesstimate economic growth much sooner than BBS releases its accounts. We can, in fact, do this exercise by properly estimating a nowcasting equation — but I am going to stop here, this is a free blog-post, and running econometric regressions is way too much work.
Data source: BBS through CEIC Asia database. Underlying spreadsheet is available upon request. I am also open to further, more rigorous analysis — please contact by email if interested.
Further reading
A study of lights at night suggests dictators lie about economic growth
The Economist, 29 Sep 2022
Coal shortage shuts 2 units of Payra, Adani powers plants
Emran Hossain, 8 Oct 2023
অর্থনৈতিক সংকটে ব্যক্তিখাতের ভোগ ও বিনিয়োগ ব্যয়ে ধস!
ইসমাইল আলী, 14 Dec 2023
Exchange rate to float within 'crawling peg' corridor
Jubair Hasan, 21 Dec 2023
খাদ্য আমদানিতে বিশ্বে তৃতীয় বাংলাদেশ: এফএও
ইফতেখার মাহমুদ, 23 Dec 2023
Economy will be better if dollar crisis and inflation contained
Zahid Hussain, 31 Dec 2023
Private sector foreign loan repayment $5.21b higher than receipt in 11 months
Sakhawat Prince, 3 Jan 2024
Ship-breaking industry saw decade's lowest imports in 2023
Mizanur Rahman Yousuf, 7 Jan 2024
Capital market makes no direct contribution on industrialisation for first time
Niaz Mahmud, 9 Jan 2024
Bangladesh offers apparel to UK way cheaper than competitors
Jasim Uddin, 10 Jan 2024
Food price paradox: How far will the effects reach?
Shadique Mahbub Islam, 12 Jan 2024
Jagaran Chakma, 19 Jan 2024
Do you think the budget problems will finally get the government to privatise some of the loss making state owned enterprises? Like the sugar and jute mills, the oil refinaries and fertiliser production plants. The government has finally legalised private companies to enter the oil refinary and internet submarine cable business. Do you think we'll see the deregulation and privatisation of electricity market markets we're seeing in the rest of Asia like Singapore, Japan and most recently Philippines?