Financer des baisses d’impôts avec du carbone

Les nominations par Donald Trump de loyalistes radicaux à des postes de cabinet ou à la tête d’agences fédérales clés doivent être interprétées comme un signe que la nouvelle administration "ne plaisante pas" quant à la mise en œuvre de son programme de campagne. Cependant, le caucus républicain au Sénat a choisi un leader plus indépendant et traditionnel avec John Thune. Cela suggère que faire adopter le paquet fiscal ne sera pas une tâche facile.

Nous analysons comment les économies réalisées grâce au recul de l’IRA – par exemple la suppression du crédit d’impôt pour les véhicules électriques – combinées aux droits de douane, pourraient combler une partie du déficit causé par l’extension de la TCJA, la réduction de l’impôt sur les sociétés et l’exemption des prestations de sécurité sociale. Il serait erroné de considérer les principales propositions économiques de l’administration Trump sur la transition verte et le commerce international comme purement idéologiques : elles contiennent aussi un aspect de "nécessité fiscale".

Les baisses d’impôts seraient, en partie, financées par des émissions de carbone supplémentaires. Si les investisseurs pensent que la nouvelle administration n’ira pas aussi loin que le programme le suggère, parce que Donald Trump voudra préserver le rallye boursier, ils doivent aussi accepter que, s’il ne revient pas sur l’IRA ni n’augmente significativement les tarifs douaniers, il ne pourra probablement pas offrir tout le "coup de fouet" fiscal qu’ils attendent.

La Réserve fédérale a déclaré qu’elle ne "spéculerait pas" sur ce que l’agenda de Trump pourrait signifier pour l’inflation, mais même avant l’investiture de la nouvelle administration, les données ne sont pas encourageantes en matière de prix à la consommation. Pour le deuxième mois consécutif, l’inflation sous-jacente a accéléré. La baisse des taux en décembre est sur le fil du rasoir.

La BCE devra aller plus vite et probablement plus loin que ce qu’attendait le Conseil des gouverneurs dans le processus de "retrait des restrictions". Une question clé actuellement à Francfort est de savoir s’il faut être transparent à ce sujet et passer aux indications prospectives. Le discours de Piero Cipollone la semaine dernière allait dans ce sens, tandis qu’Isabel Schnabel préfère s’en tenir à une approche purement dépendante des données. Nous pensons que les forces gravitationnelles penchent désormais en faveur des colombes.

How to talk to a fiscal hawk

We think that the - close - choice by the Republican caucus in the Senate of John Thune as new majority leader is an important signal. He is a “traditional Republican”, who has long maintained a certain distance from Donald Trump and the “MAGA“ version of conservatism. In 2020, he was among those who accepted the result of the elections. Six Republican Senators objected then to the electoral college count, including Rick Scott, who was seen as the “Trumpist candidate” to the leadership this year and was defeated in the first round of the vote in the Republican caucus. True, Thune has narrowed much of that distance recently, and initially offered to expedite the vetting process for some presidential appointments, but his instincts would make him a good candidate to organise the stalling of some of the most spendthrift projects of the Trump administration. He supported the Tax Cuts and Jobs Act (TCJA) in 2017, but there was less focus on public debt dynamics at the time, and as we discussed last week, the additional tax cuts in the platform (corporate rate down to 15% from 21% and full exemption of social security benefits) are massive.

This adds to our belief that the fiscal projects may be the most fragile part of the new administration. There is clearly an awareness around the President-elect of the need to reassure Congress on the deficit front, and this is how we understand Elon Musk’s task in the cabinet, where he intends (in his own words) to cut up to 2 USD trn of “wasteful government spending”. This would be a huge task. According to the Congressional Budget Office (CBO)’s tables, total federal spending stood at USD 6.1tn in 2023, out of which interest payments stand at 0.6tn. This leaves 5.5tn to “play with”, but the major welfare state programmes – Medicaid, Medicare, and Social Security – already absorbed USD 3.8tn. This leaves 1.7tn of “discretionary spending”, nearly half of which is directed to defence, which the Republican party is unlikely to touch. During his campaign, Donald Trump indicated he would not touch the parameters of social security. There could be some savings on healthcare spending without changing the parameters of the benefits by reducing drastically the cost of medications. This avenue was tried under Trump 1.0 but failed, partly because of the efforts of the Republican party itself. Savings on operational costs may not be that fruitful. For instance, the administration cost of social security amounted last year to only USD 7bn (equivalent to 0.5% of total benefits paid).

A reform of social security is however necessary. For European readers, its financial engineering is quirky. In principle, the tax revenue earmarked to social security – levied on wages and benefits themselves - exceeds the overall amount of pension paid. This creates a surplus, which is invested in non-marketable Treasury securities paid at market rates (the social security “trust fund”). If the sum of the payroll tax (USD 1,233 in 2023), interests earned (USD 67bn) and income tax levied on social security benefits (USD51bn) is smaller than the pension outlays, the trust fund is gradually depleted. This point has been reached, with a decline in the trust fund’s assets of USD41bn last year, from a 22bn loss in 2022. ​ The latest annual report of the trust fund suggests that, if nothing changes, all the assets will be consumed by 2035 (see here). Of course, exempting social security benefits from income tax will hasten the deterioration of the system.

The social security system, created in 1935, is seldom reformed. The last time – because under the parameters of the time, trust fund depletion was right on the corner – was in 1983. It was Alan Greenspan who was in charge of designing a solution, and he had to find a compromise between a House of Representatives held by the Democrats and a Republican Senate (ultimately, a rise in payroll taxes and part-taxation of social security benefits combined with a gradual increase in the retirement age). Even if Ronald Reagan was often portrayed as quite extreme at the time, these were much more peaceful days politically than today, with a capacity to cut bi-partisan compromises which looks completely out of reach in the current configuration. It is such a sensitive issue that no single-party solution is realistic in our view, especially given the very small majority margin the Republicans have secured in the House.

Where the Trump administration can more easily offer some interesting concessions to Congress to get its tax projects through, it is on green spending. Throughout the campaign, the President-elect made no mystery of his determination to terminate the Inflation Reduction Act (IRA)’s spending earmarked to the transition to net zero. In our exploration of the IRA two years ago we made the point that, for all the envy the IRA has drawn from European observers, this remains an incredibly expensive programme. A major flaw in our view is the fact that on many aspects, there was no cap on future spending. The “scoring” of the tax credits by the CBO at the time of the congressional discussion of IRA (USD 14bn for the clean vehicle incentive) was very static – i.e. did not take on board the change in behaviour the incentives would create. Using the macroeconomic model REGEN, which precisely takes on board a shift in consumers’ preferences towards electric vehicles in response to the USD 7,500 tax credit, the Brookings – which cannot be accused of systemic climate-scepticism – puts the cumulative price tag to USD390bn by 2031 as the share of EVs in new vehicle sales would reach 44% (see link here). Of course, estimating the dynamic effects can be acrobatic, but the Brookings’ findings are very similar to those of the Penn-Wharton Budget Model (USD393bn by 2032, see the link here). The Brookings and the Penn-Wharton differ on the magnitude of the fiscal cost of the other electrification programmes (USD263bn by 2032 for the latter, another USD390bn for the former) but in any case, this could go a long way to pay for the cut in the corporate tax rate and extending the business part of TCJA beyond 2025 (see Exhibit 1).

Now, this would plug only part of the additional hole in US public finance which the Trump platform would drill (there would remain 2.9 trn to cover by 2031 to pay for the extension of the household part of TCJA and the exemption of social security benefits from income tax, using the Penn-Wharton estimates (see here), but we do not see how, from a purely financial point of view, the Trump administration could renege on its pledge to scrap at least significant parts of IRA, especially since, in terms of political dynamics, Republican fiscal hawks also tend to be sceptical on the green transition, so that convincing them to take a demise of the IRA in consideration should not be too difficult.

Some observers have put faith in Elon Musk’s proximity to Donald Trump to dampen his general allergy to the green transition. However, according to Reuters, Tesla has declared itself in favour of scrapping the tax credit – Elon Musk himself has tweeted in favour of scrapping it – and to some extent it might be Tesla’s interest. Indeed, the extended tax credit helped newcomers to the EV market, while Tesla, as a pioneer, is further away on its marginal cost curve. Interestingly, the equity market reaction after the Reuters news came out penalised Rivian – a competitor to Tesla – more than Tesla (-8.9% versus -3.3%).

Trade tariff income would be another way to plug the hole. A back of the envelope calculation would estimate the potential increase in customs duties collected in the US to roughly 150bn annually on the basis of a blanket 10% tariff (a tripling from the current level), amounting to c. 1 trillion by 2031 (although this does not take into account the feedback loop from, potentially, some of the imports being replaced by domestic production, thus yielding different income for the government). A 10% blanket tariff, especially if combined with a 60% duty on Chinese products, although explicit in the Trump platform, would be at the upper end of what is probably in the pipeline given the likely disruptions to the US own production framework. In addition, such naïve calculation fails to take on board the adverse effect the tariffs would have on other sources of government income by lowering economic growth. But we think that it would be a mistake to consider the Trump administration’s key economic proposals on the green transition and international trade as purely ideologically motivated. Ideology matters, of course, there is also a “fiscal expediency” aspect to them which we think is somewhat lost in the current debate on the extent to which Trumpnomics 2.0 will be implemented. Take out the removal of the EV tax credit and the hike in customs duties from the frame, and the chances to get the Republican agenda on tax cuts through would significantly fall. This must be kept in mind when hearing all those who have been arguing that “Donald Trump will mellow”. In effect, the project consists in paying for tax cuts by allowing more carbon emissions.

This gets us to a very simple point when it comes to the market reaction: if investors believe that the new administration will not go as far as the platform implies, because Donald Trump will want to preserve the equity rally (there was a lengthy weekend piece on Bloomberg making this point), then they also have to accept that, should he not roll back on the IRA nor hike tariffs significantly, he probably won’t be able to give them all the “sugar rush” they expect in terms of tax cuts ahead. Note that in any case, even when taking a very generous basis for the tariff’s income, combined with scrapping the electrification tax credits from the IRA, the entirety of the tax cuts platform could not be funded. ​

Halting the green aspects of IRA will likely put a brake on American decarbonation, together with the appointments – to be confirmed by Congress – of Lee Zeldin, a Republican Congressman with a poor voting record on environmental issues as head of the Environment Protection Agency (EPA) and Chris Wright, CEO of a fracking company, as Energy Secretary. Optimists will want to console themselves with the fact that the decline in the carbon-intensity of the US economy (here GHG per unit of GDP) has been declining at a very steady pace since the mid-1990s, without any visible pause under Trump’s first term, as if the various federal policies had little impact. A myriad of decentralised decisions by businesses concur to this decarbonation, and state regulation in key parts of the US territory will probably continue to push in the same direction. Yet, what Exhibit 2 also suggests is that despite the halving of carbon intensity, the sheer volume of carbon emissions has only modestly fallen from a peak in 2007. An acceleration of decarbonation is necessary, and it would take massive optimism to argue that this will be outcome of the looming presidency.

Inflation before the inflation shock

As all eyes are turning to the Senate as a potential “Trump dampener”, the Federal Reserve behaviour will also be key. We discussed last week Jay Powell’s refusal to engage in precise forward guidance on the pace and magnitude of the future removal of monetary restriction. This was probably to some extent driven by the Fed’s ignorance of the timing and substance of the Trump’s administration projects, but a not-so-straightforward dataflow explained some of that prudence as well. Last week’s data releases vindicate this approach. Indeed, the Consumer Price Index (CPI) print for October came out in line with expectations, but without breaking the “resistance line” above 3% yoy for core (3.3% in October, same as in September, 3.2% in August). When looking at the momentum (see Exhibit 3), the re-acceleration observed in September continued, including for the CPI component which the Fed tends to focus on (services excluding rents). In the commentariat last week, much was said on the outsized contribution from a few items (e.g. used cars) but here we follow the advice of Ethan Harris, former Chief economist of Bank of America (and your humble servant’s former boss in another life) who in his column last week invited everyone to look at alternative measures of underlying inflation, such as the Cleveland Fed’s median and trimmed CPI (see Exhibit 4). These measures were precisely developed to attenuate or even mute the impact of “outliers”. Even with this perspective, the diagnostic of a “resistance line” holds. ​

We wrote last week that another Fed cut in December was still probably the “natural slope” but also that it would not take much to send the FOMC in pause mode. There will be another CPI print out before the FOMC meets on 18 December, as well as one Employment Report (which will be much more interesting that the last one which was overly disturbed by the hurricane and strikes). If these two prints continue to paint a “sticky inflation combined with robust labour market” picture, a pause will definitely be on the cards. Donald Trump could be presented with an “uncooperative Fed” (from his point of view) even before his inauguration.

ECB hawks and doves discussing messaging

With key national governments “otherwise engaged” at the moment, the ECB is even more than usual at the centre of the macroeconomic response of the Euro area to the mounting external challenges. We took as encouraging the fact that the decision to cut again in October was made on a unanimous basis, since it meant that even the hawks are starting to take the measure of the downside risks. ​ Yet, we also believe that the impact of the cut(s) on economic confidence would be higher if the ECB made the “direction of travel” plain, i.e. resorted to forward guidance. This issue seems to be quite hot in the current discussions at the ECB board, judging by two speeches last week, one delivered by Piero Cipollone, the other by Isabel Schnabel.

Cipollone was invited to discuss productivity trends but focused on monetary policy in connection with the issue. He acknowledged the disappointing recent trend in European productivity and put the impact of monetary policy restriction (through in particular the investment channel) among the potential suspects. His point on “further transmission of our less restrictive stance will also depend on confidence that we will gradually but surely lower our policy rates further toward neutral level”. While this was followed in the following sentence by a reminder that the central bank follows a “meeting by meeting approach”, we think it is clear that he advocates a clearer path ahead. He also mentioned the possibility that the path towards more neutral rates should take on board the fact that transmission can be impaired by tight credit conditions and “signs of deterioration in banks’ asset quality”. We take this as a signal that the central bank should not take a too leisurely approach to restriction removal. Pushing the interpretation further, arguably this could mean that, should transmission be impaired, a proper “neutral stance” for the economy at large could make it necessary to bring the policy rate itself squarely in accommodative territory. In any case, Cipollone used his speech to communicate his view that the ECB should not wait until the economy has properly slowed down further before taking the necessary steps: “it could be self-defeating to tolerate an economy running persistently below potential as an insurance against possible future inflationary shocks. Such a tactic could lower potential growth, thereby weakening the economy’s resilience to both demand and supply shocks”.

Not everyone at the board however is convinced that the ECB should take such pre-emptive line. Isabel Schnabel, in her review of central banks’ instruments, finished on forward guidance. Her point of view is clear: “in a high-volatility world (…), forward guidance almost mechanically implies that central banks either take the risk of falling behind the curve or take a course of action that is inconsistent with their previous guidance”. Now, the gap between her and Cipollone may not be that dramatic (he agrees that the precise pace of restriction removal should continue to be determined by the dataflow), but we find it interesting that Schnabel, often the “leader of the hawks” at the board, felt the need to lay out the drawbacks of forward guidance: there is probably no surer sign that it is actually not so far around the corner. Some of the concessions she made at the end of the speech – mentioning “Delphic forward guidance” - suggests that she knows she is fighting a rearguard battle. Forward guidance will come, the hawks will merely endeavour to make it as vague as possible.