My article published in ‘Les Échos’ on November 22, 2024
The profitability of retail banks is in question. Is it inevitably low in France? Some food for thought. Retail banking has two engines. If one of the two does not work, profitability is compromised. Asset-liability management, i.e. interest rate and liquidity risk management, is crucial because income from loans and deposits – the Net Interest Margin – depends on the yield curve and its evolution. Like any good merchant, in this case money, the bank must buy a little cheaper than it sells. However, it borrows customers’ savings mostly on a short-term rate basis and, in France, lends mostly on a medium-long term fixed rate basis. The evolution of the spread between long and short rates is therefore an essential parameter. The bank itself takes the interest rate risk by allowing individuals, as well as small and medium-sized companies, to be not subject to it. The bank must therefore properly measure, anticipate and manage this risk, which has been successful to a varying degree for French banking groups in recent years with the sudden return of inflation and the subsequent rise in rates.
But retail banking must also be efficient with its second engine: its commercial capacity. The digital revolution has been a game changer for some years now. Marketing or the offer is one aspect. But since banking products and services are easily and quickly copied and highly regulated, there is little difference between them. On the other hand, the customer approach, the organisation of the sales force, similarly its management method, the choices of combination between physical presence and “online” banking, are key elements in the greater or lesser success of retail banks.
The digital revolution in banking can lead to a drastic reduction in the number of branches, with the idea that banking will increasingly become online without client advisors. This is a difficult path in France, as banks are much more relational (advice) than purely transactional (simple daily operations: transfers, balance checks, etc.). To make a purely online bank profitable, it is necessary to succeed in sufficiently equippingcustomers with a wider range of products and services. Those who gradually manage to do so therefore need more and more advisors and to move away from a purely “online” model. In a different way, it is possible, while streamlining its network without systematically reducing it, to use digital technology to simultaneously improve customer comfort and free up sales time in the branch for more proactivity, as well as added value provided to customers, by advisors. On the other hand, not significantly changing anything, while ignoring the powerful effects of the technological revolution, cannot serve as a solution. The time of the customer in contact with his bank mainly dueto his visiting a branch is long gone. The operating coefficient (expenses on income) would then inevitably rise to the point of suffocating the bank. With an unavoidable drop in results as a result.
Retail banking responds to the immediacy of time, through its payment services, but also – and this is probably its deepest essence – to the long term, through its advice. It supports customers in their life and business projects, which are all prepared and unfold over time. This advice requires savings, credit and insurance, all products that continueover time and require support, regardless of the type of customer. Digital technology must therefore be used to ensure customer comfort and to maximise advice time and its added value. It is a question of strategy, means and incentive systems.
The future of retail banking therefore requires an appropriate financial and distribution policy, excellent operational control of these two drivers and the ability to give meaning and strong motivation to the human resources on which any service industry ultimately relies. Retail banking therefore has a future.
Olivier Klein Chief Executive Officer of Lazard Frères Banque Professor of Economics at HEC
Today, the urgency of change in the management of public administrations is essential. For companies as well as for public administrations, without confusing the two, changes in the behaviour of employees and users or customers or technological revolutions often require deep transformations to survive and develop for some and to remain effective and legitimate for others. In a changing world, nothing is a given. And, without anticipation, crises lie in wait, which then force sudden, uncertain and socially painful ruptures.
We have to be constantly attentive to changes in the conditions in which we carry out our activity, regularly rethink the validity of our model, while maintaining a methodological doubt so as to never be caught up in our own certainties. At the same time, to avoid Brownian movements, it is essential to rely on a clear analysis of what is invariable in our activity. What it is fundamentally and how it is sustainably useful to people and the economy. Thus, clearly conceiving the very essence of our activity and simultaneously perceiving the changes linked to our mode of practice is a crucial key to forging a good strategy and reaching your goal in the best way possible, ensuring a calm transformation and not a brutal disruption.
But as necessary as this may be, to succeed, it also requires thoughtful and appropriately organised change management. We have to anticipate the reactions that employees, customers/users, and even the competition will manifest in response to the changes that we want to implement. To anticipate correctly is to allow ourselves to act correctly.
Consistency is also a fundamental key to success. It must be totally implemented in the strategy. Otherwise, there is a loss of meaning and with an inconsistent course and divergent directions given, we will go nowhere. But there is more. There must be constant alignment of the strategy, the means needed to achieve it and the incentive systems. When the means implemented are in line with a relevant strategy and the incentive system ensures that each person or team is inclined to direct their action towards the achievement of the proposed strategy, success is often the result.
Finally, it is necessary to have adequate change management and support. This means detecting and considering obstacles to change in advance, thanks to feedback from the teams themselves, especially because they are stakeholders in the change but also because they are in their field. Change cannot only be driven from above. It must be the result of ongoing dialogue between managers and managees. This process is demanding but necessary and fruitful.
Which logically leads to a well-conducted trial-and-error process. Change must have a clearly defined course. But, if it is necessarily a well-thought-out process, it must be flexible. Change should not be planned rigidly and stuck to no matter what. Very flexible and dynamic planning, with a clear route, integrating the encountered realities, in a back and forth between the conceptualisation of the process followed and the reality that is revealed as it is deployed, allows one to achieve one’s objective with much more certainty.
Life is change, a permanent evolution. Just like companies and administrations and their environment. We must therefore think about and lead the essential changes, before being forced to do so by the otherwise inevitable crises. Companies are born and die when they have not been able to adapt. Public administrations do not die by themselves, but they can experience such entropy that they become less and less efficient and more and more costly, and can even lose their legitimacy. This then leads to deficits and debts which, when facing the wall, can lead to a disruption that is always hazardous and painful.
Olivier Klein Chief Executive Officer of Lazard Frères Banque Professor of Economics at HEC
To fully understand this, it seems necessary to make a clear distinction between the quantitative easing policy of central banks consisting of massively supplying banks and markets with “liquidity”, to be more precise in the money of issuing institutions, prolonging and broadening their historical action of last resort, and that consisting of the purchase of assets on the financial markets, public and private bonds, or even stocks. These two ways of conducting a QE policy induce the same inflation of the assets and liabilities of central banks, but by different means. In the first case the monetary authorities lend to banks, in very unusual circumstances, in order to avoid a liquidity shortage in the banking system. In the second, in equally unusual circumstances they buy assets from non-banking agents who, by selling them to central banks, receive money that they deposit in the banks; the latter thus ultimately holding more central bank money in their accounts with issuing institutions. We could call the first a policy of the liabilities of the balance sheet of central banks and the second an asset policy.
But the most fundamental distinction lies in the circumstances leading to the use of these policies. In the United States, as in Europe for example, they were initially used to deal with the very serious financial and economic crisis of 2008-2009. The risk of a chain of bank failures, as well as market dislocation, required breaking these catastrophic dynamics by temporarily suspending the logic of the financial markets and the contagious distrust between the banks themselves. Which could also have led to a destructive crisis of confidence among households in their own banks. Thus, the central banks increased their balance sheets to provide the banks with the necessary liquidity. De facto, with the interbank market frozen, they interposed their balance sheets even in the exchanges of liquidity between banks. Banks with excess liquidity no longer lent to other banks and kept their central bank money in deposits at the central bank, with the issuing institution itself then lending to banks requiring liquidity. This operation of credit thus adds liquidity in the banking system and inflates the Central Bank’s balance sheet. This is how central banks were once again the lenders of last resort to the financial system from 2008. But it was also in 2008 that the Fed decided to buy “toxic” assets to prevent the loss of confidence and bankruptcy of those who held them. And also to prevent the collapse of the price of these assets, the consequences of which would have been potentially disastrous for the economy. This is how central banks were able to curb the systemic risk that was developing
very quickly and which would otherwise have led to catastrophic economic and social consequences. The same was true around March 2020, during the very severe financial flash crisis due to COVID and lockdowns. Central banks then bought many assets, including high yield assets, particularly from non-banking financial institutions, including troubled investment funds, which made it possible to very quickly extinguish the fire that was suddenly taking hold.
Quantitative easing was then used and perpetuated with a completely different objective. The violent financial crisis was over, and as a result, the economy was very slow and inflation was at extremely low levels. With interest rates close to their effective lower bound, the weapon of key interest rates in conventional policy had become ineffective. This is why central banks began to buy financial assets to stimulate the economy and try to raise inflation. Then in 2020, with the economic consequences of the pandemic and lockdowns, they did the same. They have notably bought public debt bonds, in order to support the government’s very important fiscal efforts.
The effectiveness of this policy may come from the very announcement of the implementation of such a decision. The OMT program, for example, announced in 2012, led to a rebound in confidence following its presentation, even though it was never implemented. But its effectiveness may also come, of course, from the possibility it offers central banks to take control of long-term interest rates and risk premiums, or at least to substantially influence them. In this way, they can encourage households and businesses to invest, or even consume, more, in particular by lowering the cost of their loans. Credit, and its twin, debt, have also gradually recovered since around 2017 in Europe. The third transmission channel was the wealth effect triggered as a result of the fall in long-term rates on the value of stocks as well as on real estate. This wealth effect has thus supported demand.
Let us stress, however, that such a policy of quantitative easing outside of a situation of financial stress, to be effective, that is to say to stimulate the economy, must require central banks to buy much more assets, that is to say to create much more central bank money, than during financial crises. However, even if it requires injecting even more liquidity to reinvigorate economic growth as well as credit growth, this policy of purchasing assets has been useful. It has notably prevented a triggering of a deflationist cycle.
However, it failed to raise inflation. It is likely that the 2% inflation target did not correspond to a rate that met the contemporary mode of economic regulation, that is, the prevailing structural conditions during the pre-COVID period. The combination of a sustained situation of globalisation, which weighed on wages and prices in developed countries, and a technological revolution that gave little room for manoeuvre in wage negotiations for low- and medium-skilled employees, helped by digitalisation and robotisation, induced the structural causes of very low inflation, perhaps around 1%. The efforts of central banks, seeking to stimulate the economy and inflation, then succeeded in increasing the growth rate, but not the level of inflation. And, in its search for an inflation target that is probably unattainable as well as with the help of a compass – the natural interest rate – that is very imprecise and conceptually questionable, monetary policy has persisted in using quantitative easing, even though GDP and credits had returned to a favorable trend. The consequence has been interest rates that have been too low for too long, that is to say, sustainably lower than growth rates. With, as a result, the rise of financial instability due to a strong growth in the indebtedness of private and public agents (in relation to GDP) and in a feedback loop with this increase in indebtedness and the very rapid and very strong rise in the value of stocks and real estate. Indeed, with long-term interest rates lower than the nominal growth rate, private and public players were encouraged to painlessly increase their debts, thus weakening their balance sheet.
And savers, or their asset managers, their pension funds or savings funds, like their life insurers, have sought to obtain a sufficient return. They have thus been encouraged to buy increasingly risky assets (drastically lowering risk premiums), increasingly long-term assets (thus taking on more and more liquidity risks), etc. The balance sheets of many economic players have thus become fragile both on the assets and liabilities side.
Furthermore, very accommodative monetary policies that persist for too long de facto increase wealth inequalities, further enriching those who already have some, and making real estate and the stock market less accessible to others. Finally, with all these effects combined, it is reasonable to think that they have ultimately contributed to the slowdown, to the languishing of the economy, by having notably allowed the development of “zombie” companies (which, with normal interest rates, would have experienced negative results).
This overall poor allocation of capital has very probably contributed to decreasing productivity gains. Furthermore, interest rates that are too low for too long may have, as in Germany for example, contributed to the rise in savings rates and not to their decline, contrary to the precepts of standard economic theory. A population with a declining demographic may indeed want to save more to prepare for its retirement, no longer being able to count sufficiently on the return on its “normal” savings. The high level of indebtedness of many players also sooner or later weighs on their investment capacity. All of these factors then lead to a structural slowdown in growth and not its support .
To conclude, it is difficult to exit quantitative easing policies once they have been used for too long. Thus the policies have developed asymmetrical characteristics which could be worrying. Among other things, this asymmetry can provide markets with free options to protect them on the downside while allowing them to play the upside with very limited risk. The consequent fragility accentuated in the financial structures of agents, both in the assets of balance sheets, which include highly valued or overval ued stocks and real estate, and in their liabilities, which have levels of debt to GDP rarely reached, rightly enco urages central banks to be very cautious in their balance sheet reduction (“quantitative tightening”). Reducing the balance sheet of issuing institutions too quickly or too intensely could indeed lead to major financial and economic crises. This is why we can consider that liquidity in central bank money will only be withdrawn – as at present – very cautiously. In fact, very probably, the opposite path will never fully reach its end. Furtherm ore, we are in an unprecedented realm since the experience of “quantitative tightening” is a historical first. The fight against the sudden return of inflation these last years has been successful thanks to conventional monetary policy effects (raising their pol icy rates) , without acting discretionarily on their balance sheet size, but driving their l ow-key reduction.
Ultimately, it seems that we can affirm, based on a retrospective analysis, that quantitative easing policies have very favorable effects when it comes to healing a violent financial and economic crisis by trying to contain systemic risk, that is to say, to prevent a catastrophic chain of events. However, if it is also useful to use them to stimulate growth and inflation, if we continue to use them, even when growth has returned, and, moreover, if the target inflation no longer corresponds to structural inflation, this may seem to entail more dangers than benefits. We have not yet seen all the economic and financial consequences that such a policy pursued for too long can induce. Let us bet that, barring new violent crises, and across cycles, central banks will seek to sustainably maintain interest rates at levels roughly equal to nominal growth rates. All things considered, and in order to be able to use it again in case of proven need, the unconventional monetary policy should remain unconventional.
Olivier Klein Chief Executive Officer Lazard Frères Banque Professor of Economics and Finance at HEC
BIBLIOGRAPHY
Exiting the ECB’s highly accommodating monetary policy : stakes and challenges – Le Blog Note d’Olivier Klein Revue d’Economie Financière – https://www.oklein.fr/en/exiting-the-ecbs-highly-accomodating-monetary-policy-stakes- and-challenges-2/
HOW CAN WE AVOID THE DEBT TRAP AFTER THE PANDEMIC? – Le Blog Note d’Olivier
KLEIN Revue d’Economie Financière – https://www.oklein.fr/en/how-can-we-avoid-the-debt-trap-after-the-pandemic/
The benefits and costs of asset purchases ECB – https://www.ecb.europa.eu/press/key/date/2024/html/ ecb.sp240528~a4f151497d.en.html
“It is in developing the employment rate, encouraging work, revaluing the value of work, social mobility, encouraging entrepreneurship, education (a decisive factor), innovation and growth that we must find solutions… And not through additional taxes on income from work or savings, any more than on businesses”
Correcting the poor trajectory of our public finances is a necessity that has become an urgent priority. To avoid a public debt crisis, ensure France’s independence and regain credibility, and therefore a real capacity for influence within the European Union. To this end, it is theoretically possible to increase taxes and social security contributions, reduce public spending and strengthen growth through structural reforms and investments for the future. However, each of these measures, in the specific situation of France, will not produce the same effect and will not have the same effectiveness. Let us focus here on increasing taxes, which might seem, at first glance, to both reduce the public deficit and improve social justice. The reality is quite different. Increasing taxes in France could worsen the vicious circle between very high redistribution – in itself and compared to similar countries – and relatively high income inequality before redistribution. By again degrading the lack of competitiveness and the inadequacy of our offer and thus reducing growth and ultimately damaging the standard of living of all and the tax base itself.
Lack of competitiveness. The rate of compulsory taxes has been on an upward trend in France for a long time, reaching more than 43% of GDP in 2023, one of the highest in the European Union with around 6 points more than the average for the eurozone. With a part of public spending that is inefficient (e.g the situation of hospitals, education, the redundancy of administrative operating costs, etc.) and also significantly higher than the European average – by around 8 to 10 points of GDP – and all this for a lesser result. This state of affairs contributes to the lack of competitiveness of our offer, which is currently the heart of the issue.
Furthermore, after redistribution, income inequality in France, measured by the ratio between the income of the wealthiest 10% and that of the least wealthy 10% or by the relative poverty rate, has not changed or has changed little for over 20 years. And is among the lowest in Europe. The Gini index of post-redistribution inequality stands at 0.298, while Germany reaches 0.303 and Spain, Italy and the United Kingdom have levels between 0.320 and 0.354. Let us also add that the share of national income after redistribution held by the wealthiest 1% in France is also one of the lowest at 7.17%, compared to 8.72% in Sweden, 10.32% in Italy or 14.35% in the United States. France in fact has one of the highest levels of redistribution in the OECD. Overall, in France, redistribution reduces the ratio between the income before redistribution of the wealthiest 10% and that of the least wealthy 10% from around 20 to 9. And this ratio increases to 3 by adding the effect of public services, by comparing for each the cost paid versus the monetary equivalent of what is received by using them. The wealthiest pay more, due to the high progressiveness of taxes. Thus, 85% of people among the poorest 30% receive more in terms of public services than they pay, compared to 57% for all people in France (INSEE study of 2023 on extended redistribution).
The marginal tax rate on household income is 55.2%, compared to 47.5% in Germany. It is higher than in Italy, Spain, the Netherlands or Belgium, for example. And the tax rate on capital income is still higher than the European average despite recent reductions that have been very useful for the French economy, which has been well documented.
Ignoring this when building economic programs will obviously lead to inadequate and dangerous proposals for the economy and ultimately for the less well-off.
Inequality of opportunity. True social justice, given the reality in France, is to tackle inequality of opportunity, which is quite high compared to the European average. And it is in developing the employment rate, encouraging work, revaluing the value of work, social mobility, encouraging entrepreneurship, education (a decisive factor), innovation and growth that we must find solutions… And not through additional taxes on income from work or savings, any more than on businesses. Increasing redistribution further and further, at the particularly high level we are at, is making the problem worse, by leading to less competitiveness, therefore less production and less growth. The overall risk is very high of causing more inequality of income before redistribution and more inequality of opportunity.
And of not improving the sustainability of our public finances, or even of deteriorating it further. French history in recent decades bears witness to this nonvirtuous loop. The economy, like income, is in fact a dynamic, not a zero-sum game. Serious long term research has shown this unambiguously.
What remains is action in the gradual reduction of public spending (well chosen and well managed) in relation to GDP, as well as structural reforms and future investments to increase the competitiveness of our offer and our growth potential, and at the same time promote social justice and restore the sustainability of our public finances. And thus sustainably protect the precious asset that is the level of income and social protection in France.
When the interest rate is equal to the growth rate, as it is today, stabilising the debt rate requires a zero primary public deficit (before interest charges on the debt). And a primary surplus is needed to reduce this debt rate. Otherwise, the public debt continues to increase and, the higher the debt rate, the greater the risk of a snowball effect.
France is far from enjoying a stabilised debt situation. With a very high public debt to GDP ratio of over 110%, and a very marked upward trend in this rate (the rate was around 20% in 1980), France is experiencing a primary deficit of between 3 and 4% of GDP, with an interest rate on the debt approximately equal to the nominal growth rate. This, without correction, will sooner or later lead to a refinancing crisis.
If the interest rate on the debt were to be higher than the nominal growth rate, due to a generalised rise in interest rates or in the spreads paid by France or because of a fall in the growth rate of the French economy, the snowball effect of the public debt would become even more significant.
It would therefore be necessary to reduce public spending by around 100 billion euros. Doing it too quickly would lead to too sharp a slowdown in growth and would be difficult to accept. Doing it too slowly would lead to a new dangerous increase in public debt, which would put the country’s solvency at risk, would very probably also slow down growth due to the fear of savers and investors thus generated, and finally would risk a financial crisis which would force adjustments to be made urgently and brutally, as in the case of Spain and Portugal, for example, during the Eurozone crisis.
Let us add that given the comparatively very high level of French public spending and compulsory contributions on GDP, it is much more economically efficient to reduce the former and not increase the latter. Reducing public spending indeed contains much less risk of slowdown, and could even promote growth, compared to increasing taxes. Also, while it seems elegant to say that the choice between reducing spending and increasing taxes is a political choice, it is certainly not relevant in terms of economic efficiency in France’s current situation.
Moreover, income inequalities after redistribution are among the lowest in Europe in France and the level of redistribution on GDP is already one of the highest. Reducing income inequalities in France is therefore not a reasonable objective, because it would go against the pursued goal by further reducing competitiveness which is already too low and an incentive to work that could be improved, and therefore an employment rate that is already insufficient. Which would go contrary to the direction of the announced objectives.
Stabilising and then reducing the French public debt rate is a sine qua non condition for the sustainability of our social protection and our standard of living. Risking hitting the debt wall by refusing structural reforms or going back on those that have been carried out would risk forcing us to implement austerity policies that socially are very costly.
Let us recall that if the United States, which has a very high public debt rate and a large primary deficit, does not have the same burning obligation to date, it is because it has benefited until now from a much higher economic dynamic than ours, from a stock market yield and interest rates higher than ours, which attracts capital from all over the world. Thus, they have, until now, had no trouble refinancing their external debt as well as their public debt. However, this will not absolve them ad vitam aeternam from having to correct their public finance trajectory as well.
The pension reform has been criticised for being poorly prepared and poorly negotiated. So be it. However, repealing it would be very financially dangerous given the fragility of our public finances and the absolute need to redress them. Any sign of a worsening situation could trigger a serious French debt financing crisis. Therefore, the repeal of this reform would be understood by savers and the markets as irresponsible. Public pension spending on GDP will already represent 14.4% in 2022 compared to 11.9% in the eurozone.
But the repeal would also be very unfavourable for the French economically (for households or businesses) and ultimately for jobs and purchasing power. The only possible ways to ensure the balance of pay-as-you-go pension schemes are to lower the level of pensions, which is obviously good neither for retirees nor for the economy. To increase social contributions, through being paid by employees, thus causing a loss of purchasing power and downward pressure on demand. Or paid by businesses, knowing that these social contributions on GDP are already 50% higher in France than in Germany and 30% higher than in the euro zone on average, which ultimately amounts to less competitiveness, fewer jobs and downward pressure on wages. Finally, the third and final solution is to adjust the length of working life according to demographic changes.
Age measures (retirement age or, better, number of annuities) adjusted of course according to the arduousness of each person’s work, are the only ones capable of reconciling the interests of current or future retirees and the search for the best growth potential for the economy, employment and purchasing power. All the more so since even today many companies cannot grow as much as they could due to a lack of skilled or unskilled labor. A reminder: in France, we had 4 contributors for 1 retiree in 1960. In 2010, there were only 1.8 contributors for 1 retiree and it will be 1.2 in 2050. At the same time, in 1958, life expectancy at retirement age was 15.6 years for women and 12.5 years for men. In 2020, it is 26.9 years and 22.4 years respectively. While the retirement age is lower today than in 1958. Life expectancy in good health after retirement has also increased considerably. In France, only around 30% of people aged 60 to 64 work, while in other eurozone countries, almost 50% do so. 57% in Germany and 68% in Sweden! All neighbouring countries have indeed raised the retirement age (65 to 67 years) for the same reasons and out of realism. As a result, we must also finally wake up to the reality principle, so that our pay-as-you-go pension scheme is not endangered by the inability to finance it. This reform even if insufficient, is a step in the right direction. It is always possible to adjust it a little, but being careful not to open Pandora’s box…
Finally, two thoughts. First of all, work is not only economically necessary, it is also most often a means of integration, socialisation and self-fulfilment. Let us therefore facilitate the work of those over 60 and encourage companies to keep them, or even hire them. The second: work is not to be shared because it would be in finite quantity. It is a static and erroneous view of the economy that leads to thinking this way. Work creates work in a dynamic where supply and demand feed each other. All empirical results confirm this.
