What causes negative interest rates?

The level of the real interest rate in the longer term is independent of monetary policy. It is related to the real rate of return on capital, which, in turn, depends on the fundamental factors determining savings and investments in the economy; the central bank may influence the real interest rate in the short- and medium-term only; at the same time, this impact is limited by global factors.

The year 2020 will surely be associated mainly with the outbreak of the coronavirus pandemic and the ensuing global recession on a scale that has not been seen for decades. The economic crisis driven by the pandemic also triggered an unprecedently strong economic policy response. It was mainly fiscal policy that provided support for economic agents. Yet, central banks considerably eased their monetary policy in order to reduce the economic cost of the pandemic and help economies to relatively quickly return to their growth paths. They resorted to many tools, also the ones which not so long ago used to be called non-conventional monetary policy tools and which have now become a standard element of monetary policy toolkit. At the same time, central banks which still had such a possibility also used the most “conventional” option: they cut the short-term nominal interest rate. As a result, real interest rates, i.e. nominal interest rates in the interbank market adjusted for current or expected inflation, are now running at a very low, often negative, level (Figure 1).

Keeping real interest rates at a negative level is criticised by certain commentators. They point to what, in their opinion, are the negative effects of low real interest rates. In particular, they suggest that they discourage saving. At the same time, they put the blame on  central banks, which are seen as being responsible for this state of affairs. Of course, the level of short-term interest rates is determined by central banks; however, they do not operate in a vacuum, there are some important rationales behind their decisions  that – in the long term – are taken to keep inflation at target. Yet, central banks do not always manage to do this – despite negative real interest rates, inflation in many economies has been running far below inflation targets for quite a long time.

Why are real interest rates negative?

Real interest rates worldwide have been systematically decreasing for several decades. Negative real interest rates could be partly attributed to cyclical factors; yet, in such a long period fundamental factors are of decisive importance. However, there is no consensus among economists as to what these factors are. First of all, in line with the mainstream economics, the real interest rate depends on productivity or the potential output growth (Laubach and Williams, 2003), which has recently declined (Figure 2). This may have been driven by, among others, fewer breakthrough innovations than in previous decades (Gordon, 2017) and a lower or even negative growth rate of labour supply due to the ageing of the population in some developed economies. The structural surplus of savings over investments (the so-called saving glut; Bernanke, 2005), resulting from inter alia increased propensity to save and reduced potential return on investment in the ageing societies of developed economies (Bielecki et al., 2020) and the pro-export policy of many emerging economies could also have a negative impact on the real interest rate. The rise in the savings rate – through a high propensity to save among the richest – may have been supported by growing income inequality (Mian et al., 2020). The decline in nominal investment demand, and thus in interest rates, could, in turn, be driven by the fall in the prices of investment goods in relation to the prices of consumer goods (Lian et al., 2019).

Moreover, the fall of real interest rate could also be caused by increased demand for safe assets, which is a factor behind the significant decline in the interest rates of assets without credit risk, i.e.  interest rates on open market operations of central banks and Treasury bond yields (Caballero et al., 2017). This is so because demand for safe assets, broadly determined by growth in the world economy, grows faster than their supply, which is related to the growth of developed economies generating safe assets.

The above explanations do not necessarily contradict each other, they may be complementary. Whatever the explanation, their relative importance is also difficult to be assessed. Obviously, the intensity of these factors is not the same in all economies. However, most of the world’s economies – including Poland – are small economies open to capital flows. In their case, the real interest rate is affected not only by domestic structural and cyclical factors, but also global factors, which are ultimately dependent on the global savings-investment balance. Thus, even if a country has a higher level of potential growth than developed economies, or does not face the ageing population problem on the same scale as other countries, global factors will still lower the real interest rate.

In recent months, apart from the structural factors responsible for their long-term downward trend, the decline in real interest rates was also driven by cyclical factors related to the sharp decline in activity caused by the COVID-19 pandemic. In such conditions, in order to support economic activity and increase the probability of achieving the inflation target in the medium term, central banks which still had the possibility to do so cut nominal interest rates. This also translated into lower real interest rates.

Yet, the influence of central banks on real interest rates is limited. First, in the longer term, the equilibrium real interest rate is determined by structural and global factors, with very limited impact of monetary policy. Second, although by setting the level of nominal interest rates central banks do have an influence on real rates in the short and medium term, the decisions of monetary authorities are not arbitrary. Nominal interest rates are set in such a way that – considering the economic and inflation outlook – the deviation of real rates from their long-term level ensures that monetary policy goals are achieved. In other words, central banks respond to changes in the economic situation (cyclical changes), but cannot impact long-term structural changes. In very simple terms, when economic activity and inflationary pressures are high, real rates should be higher than the equilibrium rate. In turn, when economic activity and inflationary pressures are low, real rates should be lower than the equilibrium rate. However, this “equilibrium rate” has been systematically declining for decades, and its estimates are subject to high uncertainty.

Thus, it was not central banks that caused the real equilibrium rate itself to drop significantly in recent decades, disregarding the difficulty of determining the actual level of this unobservable variable. As a result of this phenomenon, central banks are getting closer to the so-called effective lower bound for nominal rates. That is why they are deciding to use other monetary policy tools, including asset purchases.

How do negative real interest rates affect the economy?

Both the theory and the results of some empirical studies indicate that a change in real interest rates affects the economy in basically the same way, regardless of whether these rates are negative or positive (Chmielewski 2020). A decrease in real interest rates primarily reduces the real costs of servicing the existing debt and lowers the cost of new borrowing, which supports consumption and investments. Public debt servicing costs are also declining, thus facilitating increased government spending or reduction of the debt burden.

Lower interest rates may also contribute to the weakening of the exchange rate, which increases the competitiveness of domestic production on the domestic and foreign markets. Finally, lowering interest rates – by increasing the present value of future earnings – supports asset price increases. As a result, the valuation of corporate and household assets increases, which may have a positive effect on investment and consumption. The increase in aggregate demand resulting from the above factors stimulates economic activity and reduces unemployment. The improved economic situation also boosts demand and wage pressure, and the depreciation of the currency translates into higher prices of imported goods. As a result, inflation should be on the rise. Although such a description of how changes in real interest rates impact the economic situation and price growth is largely simplified, the main conclusion that may be drawn from it is certainly justified: a reduction in real rates, in principle, supports economic recovery and boosts inflation.

But what about the “undesirable” effects of maintaining negative real interest rates mentioned by the critics? In particular, they point out that negative real interest rates make it difficult to accumulate domestic savings, which are necessary to finance investment and ensure sustainable long-term economic growth. The link between interest rates and savings, however, is not all that obvious. Of course, it cannot be questioned that when real interest rates are negative the real interest rate on deposits and term deposits also usually falls below zero, which means that holding assets in this form becomes less attractive. But there is also the other side of the coin, as the real interest rate on deposits is not the only factor determining savings. Simply put, the impact of the interest rate on savings depends on the relative strength of the substitution effect and the income effect, which affect savings in opposite directions. The substitution effect is driven by the budgetary constraint: the higher the interest rate, the higher the savings rate – we put off more consumption for the future, as it becomes relatively “cheaper” than current consumption. However, since savings are non-consumed income, their amount also depends on the level of income, i.e. the amount of the budget constraint itself. This constraint is usually weaker at lower rates, because lower rates, stimulating the economic situation, increase income from labour and from the operating surplus, although they obviously reduce income from property. In less affluent societies, where property income plays a relatively smaller role, the effect of a rise in labour income usually outweighs a decrease in property income. Therefore, since the income effect of lowering interest rates is positive, then with the same savings rate we can save relatively more. Furthermore, with higher income the savings rate also increases due to the decreasing marginal propensity to consume.

The sharp decline in real interest rates in the recent period resulted from cyclical factors and was aimed at limiting the negative impact of the pandemic on economic activity and price growth. As the results of analyses indicate – for example for Poland – this goal has been achieved.

So much for the theory, and in very simplified terms, because other factors, such as structural conditions, fiscal policy, etc. should also be factored in. As the theory does not give a clear answer, the impact of the interest rate on savings is an empirical question and depends on the specific situation. Unfortunately, the results of empirical studies are also not unambiguous in determining the direction and strength of the relationship between the interest rate and savings, especially in the scale of the entire economy. In Poland, the bulk of savings are not created in the household sector, but in the sector of non-financial enterprises, whose profits do not depend so much on the level of interest rates, but much more on the economic situation, which benefits from low rates. As a result, the findings of the studies differ depending on the definition of the savings rate, the choice of the sample or the adopted research methodology.

The sharp decline in real interest rates in the recent period resulted from cyclical factors and was aimed at limiting the negative impact of the pandemic on economic activity and price growth. As the results of analyses indicate – for example for Poland – this goal has been achieved (see NBP’s November projection of inflation and GDP). This, in turn, means significant benefits for savers – thanks to the reduction of the negative impact of the pandemic on the economic situation, the growth in wages and income from economic activity is relatively higher. The real rate of return on selected forms of investment is of course still important, but without income it would not be possible to save at all. If, on the other hand, in such circumstances, i.e. when structural factors and the current phase of the business cycle justify keeping real interest rates at a negative level, we would strived to keep these rates above zero, we would incur significant economic costs. Economic activity would be lower, which would boost unemployment and worsen the financial situation of many companies and households. The continuing decrease in activity would also lead to a decline in inflation and would create the risk of a deflationary trap. It is not difficult to guess that in such circumstances the possibility of accumulating savings would be limited, and probably for a longer period of time. In this context, it should be noted that currently there is a high foreign trade surplus in Poland, therefore it does not face the problem of too low savings, but rather too low investment.

Keeping real interest rates at a negative level in the current economic situation is conducive to limiting the negative impact of the pandemic on the economy, including the labour market, and thus positively affects household income and savings.

For obvious reasons, it is easy to understand the dissatisfaction of people holding funds on bank deposits and who have got used to positive real interest rates, i.e. gaining profits without risk. The negative real interest rate is not so crucial as far as smaller amounts are concerned, but it can be felt in the case of larger deposits. However, it should be remembered that, to put it simply, interest on bank deposits comes from the interest paid by borrowers. Current interest rates are particularly low in order to support the economy and borrowers during the pandemic, but this inevitably means that deposit rates were also bound to fall. However, irrespective of the current phase of the business cycle, we should also bear in mind the long-term global downward trend in real interest rates. In recent years, it was possible to derive real positive income from interest on bank deposits without any risk. It is uncertain whether this will be the case in the future. However, we should remember that bank deposits are only one of the many available forms of keeping accumulated savings. As a rule, the real rate of return on many other assets remains positive over the long term. Therefore, the low level of real interest rates does not prevent the achievement of positive real returns on invested assets, although for this purpose it is necessary to invest them and accept higher risks.

The increased propensity to invest long-term and take risks should have a positive impact on economic activity, including investments, in the long term. It is therefore a desirable phenomenon, especially for economies recovering from recession. At the same time, the long-term search for assets with a positive rate of return may generate the risk of growing price bubbles in the asset markets and – if combined with excessive credit growth – could increase the so-called systemic risk. However, raising interest rates would not be an appropriate response to these risks. The negative impact of such action on economic activity and price stability would significantly outweigh the benefits of strengthening financial stability (inter alia Svensson, 2017; Schularick et al., 2020). The use of macroprudential policy tools is more effective and less costly in this case. Monetary policy and macroprudential policy are different policies with different goals, tools, and in many countries also different decision-makers. The use of macroprudential tools makes it possible to influence those market segments which generate imbalances at a significantly lower cost for economic activity than a potential response of interest rate policy (Svensson, 2019).

The views presented in this article are the author’s opinions and do not present the views of NBP.