Choose your country :
Choose your country ?
Italian budget discussions are going to intensify as the 27th of September approaches, date of publication of the document outlining the new budget law and when there should be more clarity on the key economic projections and deficit targets. In these focus we make a few scenarios on budget deficit in order to understand the level of flexibility available and to link these different scenarios to recent trend in GDP growth and to what they could imply in terms of spread levels offered by Italian debt. Our findings show that current spread levels already imply an easier fiscal stance, including some of the flagship measures of the government plan. We also analysed the technical picture of Italian sovereign bonds in terms of perspective supply over the next months and investigated recent trends on the demand side. under this respect and together with structural trends in funding cost, average maturity and sensitivity to the rate risk we found some supportive notes. Relative value is investigated too, versus other sovereign bonds and periphery corporate bonds, showing limited impact so far from higher Italian spreads on other fixed income segments.
Italy, our short-term view on the economy: decelerating growth, low inflation and rising interest rates make an explosive mix for public finances which means that a primary surplus is an absolute necessity for stabilising the debt trajectory
We revised our projections for GDP growth downwards to 1.1% in 2018, gradually decelerating towards 2019- 2020 potential of just below 1%. Our revision is based on a lower contribution to growth by Investments and Personal Consumption compared to 2017 as under current law, their growth dynamics are expected to moderate.
Still one month to go until we have clarity on the Budget Law but growth and political development are increasingly intertwined
Lower growth could have a significant impact on the political front as it could exacerbate the contrast between the M5S and League coalition and the Finance Minister and could also increase friction between M5S and League on how to implement certain fiscal expansion measures (universal income vs flat tax), should the decision be to keep it contained.
Currently, we expect that both M5S and League will temporarily overcome their programmatic differences and show some pragmatism about the implementation of the fiscal expansion package (i.e. limited fiscal expansion), with their minds focused on the May 2019 European Elections as their new political goal. In the meantime, we expect rhetoric to become occasionally confrontational with the EU, with a diplomatic role for the Finance Minister to keep dialogue open. As per recent news, though, the risk that deficit could hit the 3% level seem to have decreased and the Government appears more inclined to adopt a gradual implementation of their program.
We should have clarity by around 27 September when key figures for GDP, deficit and debt projections will be announced. Other key events will be:
By 15 October, the 2019 Draft Budget will be submitted to the EU Commission. By 20 October, the government should submit the Budget Law to Parliament for debate. Between end of August and end of October, Fitch 31/08 (BBB outlook stable), Moody’s end of October (Baa2, outlook negative) and S&P 26/10 (BBB outlook stable) sovereign rating review.
So far, little clarity and many assumptions. But exactly how flexible are Italian finances? Walking a tightrope
Key public finance indicators as presented by the previous government as targets for 2018 and 2019 are likely to be difficult to achieve due to a weaker macroeconomic performance than assumed when the budget was presented. In addition, neutralisation of the VAT hike would reduce fiscal space by a further EUR 12.4bn.
Many assumptions have been made over the past month and seeking to determine the degree of flexibility available, we provide a few examples here.
Case 1: implementation of very minimal program, including only the very initial implementation of a flat tax for small businesses and VAT hike neutralization, funded by a mix of savings and revenues. This scenario would be comparable with that reported in newspapers targeting a deficit of 1.8% and compatible with the implementation of only a few fiscal measures set out in the government contract.
Case 2: partial initial implementation of certain flagship measures proposed by both League and M5S, funded by a mix of savings and revenues and with some internal rationalization in terms of tax expenditure (e.g. implementing some form of universal income/support for people on low incomes but cutting the EUR 80 support introduced by the previous government). This scenario would be just a little more parsimonious than that apparently priced in by the BTP-Bund spread (according to a Bloomberg survey, the current spread is aligned with a 2.7% deficit).
Case 3: as a benchmark, we consider the full implementation in one go, using estimates of all measures introduced in the government contract.
In Case 1 and 2, we assume that the government will decide against deficit financing for all measures and seek to source as much as possible from a combination of measures for up to EUR 20bn. Case 3 would require deficit financing.
The government could look at the possibility of funding the program using a mix of measures, such as fighting tax evasion (which accounted for EUR 20bn in 2017, EUR 16bn on average over the past 5 years), reducing the cost of politics, a tax amnesty and reducing tax expenditures. It is vital that a mix of tax evasion revenues and proceeds from other sources be found or the deficit will rapidly fall to the 3% range or below, even in Case 1. With this budget, the government is clearly walking a tightrope.
There is also hope of finding fiscal space by negotiating flexibility with the EU commission. Should the Finance Minister succeed in obtaining some flexibility (e.g. not having to introduce the planned change in the cyclically adjusted deficit of approx. 0.6% of GDP), this would generate up to an additional EUR 10bn which could be used to partly cover the VAT hike neutralisation. But this outcome would only be likely, albeit partially, on the assumption of a sensible budget to start with (i.e. not viable in Case 3). We do not take this additional space into account in our computations. Furthermore, in our estimations, we do not take into account possible positive spillovers on short-run GDP growth of the fiscal package.
Please note that in the absence of any clarity from the government, the cases below represent possible combinations to define a few cardinal points for reference purposes, given the uncertainty hanging over the budget law.
The debt sustainability matrix is a tool that can be used to gain an effective understanding of how different budget proposals will work out
The matrix also helps us understand that the debt-to-GDP ratio would decline in the event of significant fiscal slippage (such as Case 3) only if nominal GDP growth were to move into the 5% to 6% range, which would mean either very high inflation of very high growth. It is very difficult to imagine that the fiscal package delivered would result in GDP growth moving into such territory ( without first affecting the level of 10y rates).
A return to BTP spread volatility in August as the moment of truth on the government budget approaches
Following some stabilization in July, BTP spreads trended higher again in August, reaching levels close to recent previous highs recorded in May in the 10-yr maturity (286 bp by mid-August vs 290 bp in May). However, the short-term bucket of the Italian yield curve proved to be more resilient to recent volatility than in May, as the 2-yr BTP-Bund spreads reached a peak of 196 bp by mid-August, much lower than 340 bp reached in May and lower than June’s 230 bp spread level. Volatility was also lower following the May spike in the short-term segment of the Italian yield curve. After having put Italy on watch at the end of May, Moody’s announcement that it was postponing any decision to the end of October pending the release of the budget details in September, gave some relief to BTP spreads in the short term and also helped to put more pressure on the government to keep it from adopting a fiscal stance that is too loose.
What sort of government budget appears to be implied by current spreads?
According to median forecasts collected in a recent Bloomberg survey of banks, it seems that current market spreads already imply a 2019 debt-to-GDP ratio of close to 2.5%/2.7%, higher than the 1.6%/1.8% threshold which, as we mentioned before, is being reported in government-backed newspapers and compatible with the implementation of only few fiscal measures in the government contract (Case 1, of a very minimal program, outlined in the previous section). Accordingly, this is not an extreme scenario but a more aggressive one than the fiscal easing which is consistent with recent statements. Based on a more in-depth analysis of the results of the survey, depending on the outcome of the budget proposals and target debt-to- GDP ratio for 2019, a sort of spread consensus has been built for the BTP spread. In the survey, a 0.8% debt-to-GDP ratio would match with a median spread level of 150 bp, a 1.8% debt-to-GDP ratio would be consistent with a spread level a few basis points higher than 200 bp, while 3% would appear consistent with a spread in the region of 300 bp. Sharp confrontation with the EU with a very aggressive deficit projection of higher than 6% would lead, according to consensus estimates, to much higher spread levels of between 450 and 500 bp. In a nutshell, therefore, a risk premium has already been attached to the outcome of the budget law and to reassuring statements made by the Finance Minister.
Limited contagion of other periphery sovereign bonds so far while periphery corporate bonds have also shown resilience vs BTPs on a relative basis, showing market confidence in ECB’s set of tools
In August, the Italy-Spain spread reached a historical high in the 7-10yr segment, while the spread was recently close to 2011 levels in the 1-3 yr segment. The spike of Italian spreads at new historical highs on long maturities vs Spanish government bonds underlines market confidence in the ECB’s capacity to contain contagion, thanks to the comprehensive set of tools introduced and built up over recent years and now fully available, together with the ongoing effects of QE and negative interest rates, despite the announced end of new purchases in December. Additionally, periphery corporate bonds remain correlated to Italian sovereign spreads but on a relative basis they have proved to be quite resilient so far compared to past experience. In this case, the CSPP has also been playing quite a significant role, together with a supportive macro and micro picture and financial conditions that are still easy.
Italy-Spain spreads in short- and long-term segment
Italian banks maintaining easy financial conditions and accelerating NPL stock reductions
Indications from the very latest survey on Bank Lending Standards conducted on Italian banks were mostly in line with signals from banks in other EZ countries. The percentage of Italian banks tightening conditions is still negative despite a slight rise and decelerating loan demand from companies. Accordingly, lending standards applied by Italian banks remained in easy territory in the July survey. At the same time, the ongoing reduction of Non-Performing Loans by Italian banks seems to be unaffected by political turmoil and higher spreads, as the very latest figures published on the stock of Italian banks’ net NPLs show a sharp drop in June to levels not seen since 2010 and which have more than halved since end-2016 peaks. Spread volatility has thus so far not affected Italian banks’ capacity to continue down this path.
The ongoing reduction of NPLs by Italian banks accelerated in June
What about sovereign debt supply over the coming months?
Looking at trends over recent years, net issuance of Italian sovereign debt has experienced a downward trend, decreasing from yearly figures of around EUR 70/80bn in 2013 and 2014 to the EUR 40/50bn area of the following three years. In 2018 the final figure is likely to be lower than last year, probably in the EUR 30bn area. In fact, as of July, year to date net issuance reached a value of EUR 56bn, lower than average cumulated numbers of the last three and five years.
Following recent peaks, net issuance should become negative over the coming months given that the seasonality of net issuance is an established fact, based on monthly data for the last five years. The following points can be made in this regard:
1) H1 always shows a rather steady rise in net issuance to levels close to each year’s peak, and then in H2 net issuance tends to fall off (this has been the case over the last 5 years and is also the trend in 2018).
2) In H2, net issuance always tends to be negative in August and December. A thin market and lower trading volumes are likely to be among the underlying causes of this seasonal occurrence as far as August is concerned. The fact that yearly funding tends to be front loaded in previous months, together with the liquidity issue are probably the reasons behind the very negative net issuance which is common in December.
Accordingly, supply pressure during the remainder of 2018 is likely to decline compared to H1. Net supply should stabilize and turn negative in H2 and the pace of gross issuance is likely to almost halve compared to first five months. These trends should reduce potential pressure from the primary market as, on the demand side, recent available data show that foreign investors reduced exposure sharply in May and June. In these two months, Italian banks increased holdings of BTPs by EUR 28bn, covering most non-resident selling. July saw a sort of stabilization in these trends as Italian banks increased their holdings only marginally compared to the previous two months, while Target2 liabilities edged down for Italy in the same month. The pace of purchases by the ECB remains in the region of EUR 4bn on a monthly basis but in the last three months of the year it is likely to halve to around EUR 2bn.
2018 monthly net issuance pattern of Italian debt very much in line with average numbers of the last three years
A look at recent supportive structural trends on funding costs, average maturity and sensitivity to the interest rate risk
Over recent years, the profile of Italian public debt has improved not only in terms of lower average cost and reduced sensitivity to the interest rate risk but also in terms of a lower short-term refinancing risk, thanks to an increase in its average maturity. Since 2011, the average cost of funding of Italian public debt has declined sharply and over the last three years has been between 0.5% and 0.7%. With a lag of about three years, this trend has finally also impacted the average cost of funding of the entire stock of public debt which fell from close to 4.0% in 2014 to the current 2.7%. A breakdown of Italian debt by year of issuance shows that almost two-thirds of outstanding debt has been issued since 2013 in years when interest rates were quite low. Over recent years, an average annual amount of around EUR 400bn in debt was refinanced with less than twothirds (around 62%) refinanced by issuing bonds and the rest by issuing bills with maturities below or close to one year. Over the last few years, together with a fall in its average cost (likely to continue for at least a few years, unless there is a further sharp increase in the spread) the stock of Italian public debt also saw a steady increase of its average maturity, which went from 6.4 years to current levels of close to 7 years. This was made possible mainly by managing the trade-off between the overall funding cost and the short-term refinancing risk in the refunding strategy that was implemented. To put it briefly, the weight of short-term bills (BOT) has almost halved since 2011, decreasing from 9.8% to 5.6%, while the weight of medium- to long-term fixed income instruments has increased from 64% to 72%. This strategy also worked in terms of optimizing the trade-off between the lower cost of funding on the one hand and exposure to floating rate/short-term instruments on the other. The component of floating rate instruments linked to short-term interest rates (mainly with short-term maturities) has halved from 24% to 12% and at the same time the weight of inflation-linked instruments has increased steadily and stabilized at around 11%.
Most recent developments in the cost of debt and future outlook
On the back of the increase in Italian bond yields and spreads since mid-May, the average refunding cost of Italian debt has increased slightly in recent months. From a low of around 0.55% at the end of April, the average interest rate paid by Italy on funding increased to 0.75% by the end of June, the latest officially released data at the time of writing. Assuming a stable scenario for current spreads in the remainder of the year, our projections point to a further increase to about 0.85% by the end of the year. Assuming a stable outlook on spreads for next year as well, our projections point to an average cost of funding rising towards 1.5%, a rate that would still be lower than the 2.77% average cost currently paid by Italy on its overall stock of outstanding debt and also lower than the average rate paid on instruments maturing next year. According to our calculations, the average coupon of the mid- and long-term fixed income instruments on bonds maturing in the next three years will be 2.7% (2019), 2.34% (2020) and 2.37% (2021) respectively. A further sharp rise in spreads would therefore be needed for a material impact on the cost of funding and thus a reversal of the favourable trend seen over recent years. The other point to be underlined is the fact that, thanks to the reduction in the weight of floating rate instruments and bills, the sensitivity of the cost of funding to a progressive rise in ECB interest rates looks relatively limited. In short, unless a 2011-like move takes place on spreads, Italian debt should still feel the positive lagged effects of interest rate falls over recent years, thanks also to its longer maturity. This is important for assessing debt sustainability in the event of a negative shock to funding costs.
Italian BTPs offer spreads which look consistent with a budget deficit close to the 2.5%/2.7% threshold and so already seem to imply a looser fiscal stance including a partial initial implementation of some of the League and M5S flagship measures, funded by a mix of savings and revenues and some internal rationalization on the tax expenditure front. Accordingly, if the most worrying Case 3 government budget scenario is avoided, short-term technical factors and relative value versus other spread products may support a less volatile pattern of spreads, especially in the short- to medium-term segment of the yield curve. In fact and despite lower ECB purchases, primary market supply should reduce its pressure in the coming months at a time when foreign investors have reduced their exposure in a market environment which sees BTPs as the only remaining providers of spread in the short maturity segment. Our analysis of longer-term trends also shows that, if a spread shock is avoided, the increase in average debt maturity over recent years should continue to support a further, though limited, decrease in the average cost of debt looking forward to 2019.
Arm-wrestling within the Italian government and between Italy and Brussels to find a solution to the Italian budget and avoid the Excessive Deficit Procedure. After Italy’s exit from a technical recession in H2 2018, economic data are still pointing to weakness, and many risks loom in the second half of this year. Weak growth prospects are making it more difficult to comply with deficit and debt targets, although some savings may come from lower spending. In any case, the 2020 budget is the most challenging test for the government.
Sergio BERTONCINI, Annalisa USARDI
CFA, Senior Economist