This week, another step was taken in the revision of our country’s GDP series following the new publication of the national accounts. This is a double review: an extraordinary review, which takes place every four years throughout the European Union, and, in addition, an ordinary review covering the last three years (2021, 2022 and 2023). The extraordinary reviews constitute a kind of reset of the economic accounts, through which technicians carry out an exhaustive analysis of the procedures, sources and techniques underlying their calculations. This means that the GDP series can be modified over a relatively long period.
Overall, the review is relevant, although a priori This may not seem to be of an unusual magnitude either. For the period 1995-2019, the accumulated growth remains more or less the same, but from 2020 onwards, growth rates have been increasing, between two and four tenths each year. To understand the scope of this, we must therefore relate this assessment to the one carried out last year, which also increased growth in 2021 and 2022. In this way, the economic recovery after the pandemic is brought forward to the beginning of 2022, which is almost 18 months earlier than the INE had initially indicated.
The fundamental objective of these reviews is to improve our knowledge of the economy, but they also serve to verify the accuracy of statistics professionals – the INE in this case – when they present their first estimates. The following graph shows the evolution of the growth rate one, two and three years after the initial growth was recorded each year. The result is still striking for some cases, for example the year 2021. Between the growth rate (5.1%) initially estimated and the one we know today (6.7%), there is a gap of 1.6 points. This is not a small magnitude, on the contrary, it is the largest gap we know since there are homogeneous records on the INE website. There will be time in the future to do a more detailed exercise, but for the moment this analysis invalidates a somewhat strange statement in the latest press release from the National Accounts Directorate, which shamelessly states:
These extraordinary regulated reviews should not be interpreted as a correction or error in the estimates, but rather as the need to incorporate new statistical sources and techniques into the accounting framework..
I would suggest a somewhat different perspective. It is normal to correct the national accounts, but their quality must be assessed according to their original accuracy and the degree of detail of the methodology. To think otherwise is to believe that the INE benefits from the infallibility of the Pope.
The extraordinary review reveals another relevant change. Nominal GDP levels are increasing significantly every year, so that our country is now somewhat richer than we thought until now. This has important implications when making international comparisons of GDP per capita or when assessing the health of public accounts (by dividing public debt or deficit by nominal GDP). The INE details in its press release the meaning of these revisions for the year 2021 and I recommend reading the technical note to interested readers. A simple way to visualize the change is shown in the following graph, which shows the increase in nominal GDP from the point of view of production. It is clear that two very different processes can be observed. From 2010 to 2021, nominal GDP increases by 0.5% and this increase is concentrated in three sectors: construction, real estate services and traditional services (commerce, transport and hospitality). On the other hand, in the years 2022 and 2023, the increase is greater (2% and 2.5%) thanks to the increase in services with higher added value (professional services, education and private health care).
Although the evolution of the last two years (2022 and 2023) is essential to understand the recent underestimation, the first change is even more important. In this case, the increase in activity in construction and real estate services stands out. In the first case, it is reasonable to think that the INE has corrected its series in the face of multiple evidence of an underestimation of construction activity. We had already dedicated two articles to this issue at the time, in January and June 2023, to highlight, among other things, the weakness of the current statistics that until now constituted the reference for quarterly estimates. Real estate services essentially include rents, whether actually paid or those charged to the owners of their homes. The new series of national accounts is in line with a fairly widespread perception that these incomes have been gaining in importance for decades and that some statistics (for example the Family Budget Survey) do not measure them correctly. Again, this is something we are targeting in January 2023.
This analysis therefore serves to introduce an important element of caution. For example, when GDP is analysed from an expenditure perspective, the new series show stronger growth in investment and household consumption. However, I am very concerned that as the details become known and more detailed analysis is carried out, most of the increase will be explained by two elements: higher investment in construction and the fact that families are spending more money on rent. This will lead to a persistent slowdown in many other household consumption items. Investment in machinery and capital goods, which have not yet returned to pre-pandemic levels, also needs to be analysed in detail.
A similar situation is emerging with regard to GDP through income. The INE has revised the magnitude of wages for all years, which allows the gap that can be expected between wages declared to the Treasury and those recorded in the National Accounts (CN) to be widened. Fortunately, this puts an end to the absurdities of negative fraud that occurred because more wages were declared to the personal income tax than to the CN. It is now clear that the growth in employment has been accompanied by a significant increase in wage income. However, the increase in capital income (gross operating surplus) has been relatively limited and is surely influenced by the increase in rents. It will therefore not be surprising that, while in a few days we know the accounts of companies and households, a very different evolution of corporate profits continues to be recorded according to the National Accounting and the records of the Tax Agency during the years 2019. .until 2022.
In conclusion, the effort made by the INE is welcome. The new estimate confirms most of the points already known at the time with the indicators available to all users with little interest. Wages, prices, the evolution of construction, investments and exports were already known. This should make the INE think in two ways. The first is whether it is adequately exploiting all the information available in the current statistical apparatus. The answer is no and this review proves it. The second is to ask what is the use of quarterly accounting that does not give an adequate signal and we have to wait several years to know what has really happened in the Spanish economy. While the necessary economic policies are becoming increasingly urgent, we have to wait years for the national accounts to tell us something reasonable. This is not a minor problem. It is easy to think of the deficit and public debt ratios, which are indisputable signals for the markets. It is more important that citizens really know the economic situation of our country.
