Negative interest rates don’t work

I have argued that economists should be very wary about advocating steeply negative nominal interest rates. This is a case of basing policy on models which assume what they need to prove. If  we assume that changes in the level of real policy rates is the primary lever influencing demand – it is unsurprising that the solution to our problems is to abolish physical cash and set rates at lower, negative, levels.

I have outlined a host of concerns around these policies, including the fact that the causal relationship between real interest rates and consumption may be very different across demographic regimes and at different levels of interest rates. The doughnut objection remains the strongest case for no longer relying on interest rates to stimulate spending. I have also argued that the impact of negative rates on the profitability of the banking sector may result in very perverse effects.

We can now start to assess these opposing perspectives by considering the experience of economies like Switzerland, which set its policy rate at -0.75% on January 15th this year. The initial findings suggest that we should indeed be very careful about expecting our models to work as assumed.

This recent speech from the SNB is very revealing (pages 4-5). Their policy of negative policy rates is deemed successful almost entirely because of its impact on the exchange rate and the differential yield on government securities in Switzerland and abroad. In fact, long-term domestic mortgage rates are slightly higher than they were at the beginning of the year. Banks have raised domestic lending rates relative to the policy rate, precisely to compensate for declining profitability.

Reflecting the intrinsic contradictions in these policies, the SNB is encouraged, for reasons of financial stability, that interest rates to households have not fallen commensurately – because if they did it would risk creating a housing bubble. Quite right too: negative policy rates may not cause financial instability, because they may not in fact succeed in lowering rates!

Those advocating steeply negative interest rates (and abolishing cash) should think again. Models that assume what you want to prove should be abandoned – start with how the world works.

There is an obvious policy prescription for a demand shortfall: make transfers payments to households. We have the empirical evidence. If we want independent central banks to do this, subject to an inflation target, start by giving them the power.

About The Author

Eric Lonergan is a macro fund manager, economist, and writer. His most recent book is Money (2nd ed) published by Routledge. He is also a supporter of Big Issue Invest (BII), the investment arm of The Big Issue, and is one of the initial limited partners in BII’s Social Enterprise Investment Fund LP. In a personal capacity, he makes direct investments in social enterprises. He also supports and advises The Empathy Museum.

12 Responses

  1. Robert Waldmann

    What is this “causal relationship between real interest rates and consumption” of which you speak ? I haven’t seen any convincing evidence that there is any such thing.

    Yes I am both agreeing with your point in the nastiest way possible and being pedantic. In particular, I am insisting on the official classification of residential construction as a kind of investment not production of an extremely durable consumption good.

    I notice that when you get to Switzerland you discuss mortgage rates which, home equity loans used to finance consumption notwithstanding, do have a lot more to do with residential investment than with consumption.

    I am also overstating my case. I think that there probably is evidence from increases in value added taxes that a real interest rate shift of 3,600 basis points can affect monthly consumption levels (I am, of course, thinking of the first Abe recession).

    But I don’t think there is evidence that piddling shifts on the order of 50 or 500 basis points have demonstrable effects on consumption. Contemporary macro models fit the fact that short term effects are definitely small (and not demonstrably positive) by introducing habit formation. However, the associated implication that long term effects of long lasting shifts must be large doesn’t show up in the data.

    Real interest rates affect residential investment (I personally also find it very hard to find an effect on non residential investment of the sign implied by theory).

    Of course Krugman wrote this long ago (and referred to it as an insight from the age of the dinosaurs ) .

    http://krugman.blogs.nytimes.com/2011/04/04/the-transmission-mechanism-for-quantitative-easing-wonkish/

    Also Keynes,In “The General Theory …” he argued that interest rates had an important effect on investment but not on consumption. He didn’t mention residential investment — his discussion clearly refers to non residential fixed capital investment and inventory investment (and he even wrote that the distinction between fixed and inventory investment was ill defined).

    My how I have managed to waste a lot of pixels making a pedantic verbal non point. In substance I absolutely agree with you. In particular, interest rates have a much larger role in macroeconomic theory than in fitting actual data.

    More importantly, the evidence that shifts in interest rates can make a major difference is based on huge shifts of hundreds and hundreds of basis points (prominently including Volcker’s two recessions and the extraordinarily sharp and brief recovery in between). Yet that vague recollection is supposed to make us excited about shifts of dozens of basis points around the time of QE announcements.

    Reply
    • Eric Lonergan

      Thanks for the comment Robert. I suspect that the sensitivity of consumption to changes in interest rates depends on the levels of interest rates to start with, the amount they change by, and the preferences of the population – which will be determined by demographic factors, the distribution of wealth, GDP per capita, how levered they are etc. One could summarise all this with the concept of ‘pent up demand for credit’ . My contention is that it is highly unlikely in the economies which are experimenting with negative rates that the demand for credit is sensitive to its price, but the propensity to save may be. Also, I fear unintended consequences.

      Reply
  2. Drago

    Well, according to the SNB speech, negative interest rates *do* work if your main objective is to lower the value of your currency in relation to other currencies. The question is : Will the overall impact on GDP growth would be a net positive? I’m not sure.

    Reply
  3. Desperately seeking alternatives to negative rates: James Saft | The World 247

    […] Others are advocating hugely disparate other policies. Macro hedge fund manager Eric Lonergan is one of those advocating direct cash transfers from central banks to individuals, an idea sometimes called “QE for the people.” Cash transfers would presumably be more likely to be recirculated in the real economy than money laid out by central banks for financial assets in QE, and thus might push inflation higher and obviate the need for negative rates. (here) […]

    Reply
  4. Desperately seeking alternatives to negative rates: James Saft - Jeffrey Lipton Barbados

    […] Others are advocating hugely disparate other policies. Macro hedge fund manager Eric Lonergan is one of those advocating direct cash transfers from central banks to individuals, an idea sometimes called “QE for the people.” Cash transfers would presumably be more likely to be recirculated in the real economy than money laid out by central banks for financial assets in QE, and thus might push inflation higher and obviate the need for negative rates. (here) […]

    Reply
  5. Desperately seeking alternatives to negative rates: James Saft | technology market

    […] Others are advocating hugely manifold other policies. Macro sidestep account manager Eric Lonergan is one of those advocating approach income transfers from executive banks to individuals, an thought infrequently called “QE for a people.” Cash transfers would presumably be some-more expected to be recirculated in a genuine economy than income laid out by central banks for financial resources in QE, and so competence pull acceleration aloft and nullify a need for disastrous rates. (here) […]

    Reply

Leave a Reply

Your email address will not be published.

* Checkbox GDPR is required

*

I agree