Work and income
I've retreated to my highlands to think a bit about the new edition of Tramonto . I'm trying to figure out which graphs to update and which not. My plan would be to leave the text exactly as it is (including the graphs), barring any glaring errors that come to light, but to include updated versions of some of them in the afterword, particularly those that refute or support the predictions made in the text. In general, it's not always easy to know in advance which graphs might be worth updating. For example, yesterday I made a note to update Figure 32 "to see how the transfer of resources from the real economy to financial income has evolved." Figure 32, I remind you, was this:
and the comment was this:
The impact on interest expenditure is significant, as can be seen in Figure 32. Starting from the divorce (highlighted by the dotted vertical line), we can observe three main phases, highlighted by the shading. In the first phase, from 1981 to the 1992 crisis, interest expenditure skyrocketed, doubling from 6 percentage points of GDP in 1981 to 12 in 1993 (remember that in the previous year, interest rates had been sharply raised in an attempt to defend the lira's parity within the EMS). Symmetrically, the primary requirement (i.e., net of interest) plummeted, going from 5 percentage points in 1981 to -3 in 1993 (a negative requirement indicates a surplus, meaning the government, net of interest, was collecting more than it was spending). The two movements offset each other, and therefore the overall requirement remained more or less stable around an average of 11 percentage points of GDP.
It is at this stage that the debt-to-GDP ratio explodes, as we saw in Figure 1, even though, net of interest, the government has become a net saver. Note: the rising dashed line and the falling dotted line in the shaded area at the center of Figure 32 are not a mere arabesque, no, they are something different: they are a distributive conflict. If in 1981, the government gave 5 percent of GDP to debt holders (in the form of interest expenditure) and 5 percent of GDP to the national community (in the form of net primary expenditure), in 1993, at the end of the conflict, the government gave 12 percent of GDP to debt holders and took 3 percent in net terms from the national community (because a government with a primary surplus means that citizens pay more in taxes than they receive for public services). And the holders of government bonds were and are mostly large financial institutions. Clear, right? Divorce is also the choice to transfer income from taxpayers to financial institutions, a choice that, in the words of its author, appears fully conscious.
The decoupling from the EMS slowed interest rate dynamics, and from 1993 to 2002, interest expenditure first stabilized and then declined. The total borrowing requirement dropped from 9 points in 1994 to 1 point in 2000, due to further increases in the primary surplus until 1996, and from 1996 onwards to a decline in interest expenditure. This dynamic favoured the reduction of the debt, which fell from 120 points in 1994 to 103 points in 2003. Meanwhile, interest expenditure stabilized at around 5 percent of GDP, and the total borrowing requirement tended to grow (with alternating phases) in line with the primary requirement, which, however, remained consistently negative (i.e., a surplus).
During my daily patrol of the territory with Rex, I thought I'd try to see how long it would take me to update this figure with data up to the current year. Indeed, subsequent revisions of the various sources have caused some minor discrepancies between the series from fifteen years ago and the versions currently available. This can be seen, for example, in the series for total public spending:
Updating the graph leads to this result:
And I can only comment on two things: the first is that a rapid reduction in the deficit like the one experienced in the post-pandemic period (from 9.4% in 2020 to 3.7% in 2024) is not unprecedented for our economy, as the reduction from 9.8% in 1993 to 3% in 1997 was equally rapid (and slightly more pronounced); the second is that not much has changed since 2010, except for the pandemic. We still have interest expenditure that, although it has returned from the historic highs of the early 1990s, is still higher than the overall deficit, which means we are in a structural state of primary surplus, that is, of value transfer to financial income. Don't be too confident when the "markets" complain about Italy's high stock of public debt! They profit from that high stock, and so their complaints shouldn't necessarily be taken at face value.
Getting back to the point, however, I don't think a graph like this adds much to our knowledge, in the sense that it doesn't show a real evolution, a "fourth phase" after the publication of Tramonto : the distribution structure established after the 1992 crisis remains essentially unchanged, albeit on a smaller scale. For this reason, in similar cases, I would tend to save space, avoiding updating graphs that are ultimately redundant.
What do you think? Are there any Sunset graphics you'd like to see updated, or that you think could add something interesting to the new edition's afterword?
This is a machine translation of a post (in Italian) written by Alberto Bagnai and published on Goofynomics at the URL https://goofynomics.blogspot.com/2026/04/lavoro-e-rendita.html on Mon, 06 Apr 2026 13:29:00 +0000. Some rights reserved under CC BY-NC-ND 3.0 license.