Friday, 11 July 2014

Creeping nationalisation

The FOMC June meeting minutes reveal an interesting discussion about the conduct of monetary policy in an era of excess reserves. The principal policy tools are to be interest on excess reserves (IOER) and the Fed's new overnight reverse repo facility (ONRRP). IOER has already become the principal tool for controlling short rates, and there was some discussion as to whether ONRRP is really needed as well. But in a world where USTs perform the same function for non-banks as reserves do for banks, it makes sense to control both.

The FOMC do at last appear to have accepted that excess reserves are here to stay for the foreseeable future. They still talk about “normalisation” of policy: some members clearly still hanker after a speedy return to the “old ways”, wanting target ranges for the Fed Funds rate still to be published and hoping that overnight reverse repos (ONRPP) will eventually be phased out. But ten years from now, when the system still has excess reserves (it will, trust me) and they are still using IOER as the principal policy tool, will they still describe elimination of excess reserves and return to Fed Funds rate targeting as “normalisation” of policy? I suppose they might. After all, desire to return to some form of gold standard still lingers in some quarters, more than forty years after Nixon suspended dollar gold convertibility. It can take a very long time for people to realise that a change billed as temporary is actually permanent.

But perhaps more importantly, it is beginning to dawn on the FOMC that this extended period of ZIRP/NIRP and excess reserves is fundamentally changing the nature of finance. The relationship between the central bank, commercial banks and the markets is being distorted in all sorts of incalculable ways. The FOMC have now caught a glimpse of where this is all heading – and they don't like what they see.

The watershed was the introduction of ONRRP. For the first time, the Fed is lending directly to non-banks. Well, ok, what it is lending is USTs, not cash – but as it is lending them in return for cash on which the Fed will pay interest, that is not so different from IOER on excess reserves. The Fed has opened the door to direct involvement in the real economy, and the FOMC has finally realised it is lost in the wilderness without a map.

ONRRP enables certain non-banks (money funds) to obtain a primary banking service directly from the Fed. To be sure, it is only deposit-taking at present, and there are clearly no plans whatsoever to allow money funds to make payments directly from their Fed deposit accounts. But money deposited at the Fed is money that is not being used to fund other activities. Admittedly this is actually the point – ONRRP is intended to remove money from circulation. But the FOMC is concerned that things could easily get out of hand. If money funds get into the habit of using the Fed as their moneybox, what are the implications both for monetary policy and for commercial banks? Would monetary policy in future be primarily concerned with influencing the behaviour of non-banks? Do commercial banks have any future if non-banks can transact directly with the Fed?

Here's the passage in the minutes where the FOMC define their fears about the future (my emphasis):
“While generally agreeing that an ON RRP facility could play an important role in the policy normalization process, participants discussed several potential unintended consequences of using such a facility and design features that could help to mitigate these consequences. Most participants expressed concerns that in times of financial stress, the facility's counterparties could shift investments toward the facility and away from financial and nonfinancial corporations, possibly causing disruptions in funding that could magnify the stress. In addition, a number of participants noted that a relatively large ON RRP facility had the potential to expand the Federal Reserve's role in financial intermediation and reshape the financial industry in ways that were difficult to anticipate. Participants discussed design features that could address these concerns, including constraints on usage either in the aggregate or by counterparty and a relatively wide spread between the ON RRP rate and the IOER rate that would help limit the facility's size. Several participants emphasized that, although the ON RRP rate would be useful in controlling short-term interest rates during normalization, they did not anticipate that such a facility would be a permanent part of the Committee's longer-run operating framework. Finally, a number of participants expressed concern about conducting monetary policy operations with nontraditional counterparties.”
The rest of the discussion in this paragraph is all about how to prevent a fundamental reshaping of the financial landscape. Maintaining a significant spread between ONRRP rates and IOER should encourage money funds to place deposits with banks rather than the Fed, since we can assume that call deposit rates would be at or above the IOER rate. And limiting usage of the facility should discourage risk-averse money funds from using the Fed as a super-safe deposit facility. ONRRP is intended as a monetary policy tool, not a mattress for scared money funds to stuff.

But it's nonetheless another step down the slippery slope that leads to the death of commercial banking, and indeed to the complete re-ordering of the financial system. We have actually been heading down this slope for a very long time, but the speed of travel has increased considerably since the financial crisis.

Commercial banks are slowly dying. The activities that were formerly profitable are either illegal, immoral or simply not profitable any more. And the core activities that society wants and needs are also unprofitable, at least if they are done in the way that society has come to expect – free-while-in-credit transaction accounts, inflation-level interest on deposits, fixed low margins on lending. Meanwhile, commercial banks face stiff competition from new competitors – not new banks, though there are some of these, mostly backed by large retail organisations – but an astonishing and ever-increasing range of mostly internet or phone-based providers of deposit-taking, lending and payments services. Unlike the new providers, banks are having to meet higher capital and liquidity requirements and comply with tighter regulations, while suffering margin squeeze because of low interest rates and a continual drain of dissatisfied customers. And they are still facing legal costs and fines for their past misbehaviour. It's a very tough world for banks at the moment.

And they are paying the price. The big investment banks are already breaking themselves up, and they will be followed in due course by the big universal and retail banks. What has not been achieved through regulation may yet be achieved through market forces.

But society still needs financial intermediation. And it seems that financial intermediation cannot operate without a government backstop. Deposit insurance is implicitly government backed, though the EU is trying to pretend it is not. Last-resort lending by central banks to distressed financial institutions is also implicitly government-backed, as are the various forms of cheap funding provided to banks by central banks in the last few years in the interests of economic recovery (the latest being the ECB's TLTROs). Bank balance sheets contain increasing quantities of government-backed assets in the form both of reserves and government debt. So do the balance sheets of non-banks such as insurance companies and pension funds, because of new regulations requiring them to hold higher quantities of safe liquid assets. Actually, so do the balance sheets of many corporations, who prefer to keep their surpluses in the form of 100% guaranteed government debt rather than in a bank deposit account where they risk a haircut in the event of bank failure. And the widespread use of USTs and other government-issued safe assets as collateral in repo markets means that non-bank deposits are effectively guaranteed by government even if they are not deposited in banks. Many forms of lending, too, are also government-backed, including export finance, small business lending and - increasingly - mortgages.

So the super-safe backstop offered to money funds by the Fed is only the latest in a long line of implicit government guarantees propping up the financial system. Far from ending government support of the financial system, the developments of recent years have actually made it MORE dependent on the state.

Markets, too, have become government-dependent. Markets watch central banks all the time, anticipating their actions and responding to their announcements. And exceptional monetary policy by central banks has impacted market functioning. QE reduced the supply of safe assets, raising their price, while the additional money flowing into markets as a result of QE blew up bubbles in various other classes of asset, both safe assets (gold, commodities, fine art and above all real estate) and high-yield assets. It is hard to say what market prices would be like now if no central bank were doing QE, and we are unlikely to find out any time soon: the US is withdrawing QE, but Japan is currently doing the largest QE programme it has ever done and the ECB may also soon be forced (reluctantly) to do some form of asset purchase programme. China has been doing yuan QE for a while, but if dollar liquidity becomes an issue it may be forced to repo out its USTs, which would reinforce the Fed's ONRRPs and make control of dollar liquidity more difficult. And of course the Swiss have been quietly controlling the Swiss franc market for ages. To prevent the Swiss franc rising, they've done the largest QE programme in the world relative to the size of their economy. 

In fact the entire financial system is becoming completely dependent on government. In our quest to make the financial system safe, we have made its very existence as a private sector function impossible. Central banks are no longer just players in a market: they dominate and control the market. And financial institutions are guaranteed, backstopped and regulated by government in every area of their business. Some parts of their business amount to public utilities, and there are growing calls for those areas to be explicitly supported or even taken over by government – I am thinking of universal payments systems, basic banking services (transaction accounts, small deposits and vanilla lending) and the money creation aspect of lending. And flows of funds around the system are routinely guaranteed by central banks, even if that means extending central bank support to non-banks. The economic cost of allowing that flow of funds to be interrupted even for a few hours or days is just too great.

For years, the FOMC members have blindly presided over creeping nationalisation of the financial system. Now they have opened their eyes, but it is too late. There is no going back now. For better or worse, the financial system is firmly wedded to government, and in particular to the US government (including its sidekick the Fed).

And that raises some interesting prospects. David Beckworth wonders whether Fed could take over all transaction and payments services: we could all have checking accounts (current accounts, in UK parlance) at the central bank. Izabella Kaminska argues that central banks should issue e-money in direct competition to the likes of Bitcoin, providing everyone with their own central bank e-money wallet and completely bypassing all commercial bank payment services as well. Personally I would like to see government provide safe savings vehicles for its citizens, recycling those funds into long-term investments in infrastructure and technology.

There are dangers, of course: state guarantees without supervision are easily abused, while over-regulation and rigid central planning suppress useful innovation. But reshaping the financial system as an agency of government offers an opportunity to repair the dislocation between the financial world and the real economy and restore the service function of finance. We should welcome it.


  1. RRP and RP have been a important part of how central banks operate for a long time. The fact that the fed is expanding counterparties on ON RRP is similar to undertaking QE. QE is similar to OMO's but with more counterparties. The current trend is just more of the same but on a larger scale.

    If the fed was to stick to monetary policy it wouldn't be purchasing any assets or lending (investment banking and commercial banking) and would target the interest rate just by affecting the money supply through transfers. It would also provide a pmt system. E-money should be part of what central banks provide because deposits and payments are systemic. How can you conduct monetary policy properly if people cant actually directly hold and transact in central bank money electronically?

    I think it comes down to monetary policy effectiveness. If monetary policy is ineffective then banks and markets will decline leading to dis-intermediation and the fed will have to take over. What is the most effective way to conduct MP? Directly interacting with people and without doing asset purchases. People spend on consumption more than current fed counterparties at zlb. Current counterparties will demand money infinitely or higher at zlb because investment returns are too low. People wont forgo consumption. Therefore expanding to people is more effective in generating growth.

  2. For those who have not worked in the banking industry, reading this article was a bit difficult. I have one question and it is real simple. If the whole point of money is simple to grease the process of supply and demand, why do so many central banks need to keep making more?

    1. Because we produce more than we consume, globally speaking - poverty is caused by unequal distribution not by absolute scarcity. Creating more money is a means of increasing consumption.

  3. " a world where USTs perform the same function for non-banks as reserves do for banks...."

    What did you have in mind here? Banks hold reserves: a) because they have to, when there are mandatory reserve requirements; b) for clearing purposes; or c) because under IOER, they offer an attractive alternative to other high quality liquid assets (HQLA). USTs do not achieve a) or b) for non-banks.

    Non-banks can of course hold USTs as HQLA, but so can banks. Without IOER, banks would hold USTs to meet their HQLA needs, not reserves. But, of course, with a IOER rate in excess of the rate on bills or GC repo, it's obviously attractive to hold reserves instead, an option not open to non-banks.

    So I'm guessing you meant that non-banks hold USTs as HQLA, in the same way that banks (but only under IOER), might hold reserves as HQLA. Is that right?

    1. Nick,

      Yes it's all about liquidity. Banks can substitute USTs for reserves: non-banks cannot. I've written about this before:

      Some non-banks do have mandatory requirements for safe liquid asset holdings - insurance companies under Solvency II, for example. There has been some discussion about whether, now it is accepting some non-bank deposits, the Fed could impose "reserve requirements" on those non-banks. The FOMC itself doesn't seem to have considered this yet, though. It would be another step down the slippery slope.

  4. The most dangerous thing that is happening here is that the Fed with their historical mission creep and over-reach, is taking out market discipline. This is very dangerous for societies. The efficient allocation of capital depends on market discipline in the intermediate and long-terms. The Fed is destroying market discipline, so many more ineffecient users of capital end up with significant funds. Once again, we will have a monstrous bubble mountain to fall down from when debt, once again, gets to big to service (especially for inneficient entities).