How the United States gets deflation and becomes the next Japan

By Edward Harrison On Aug 2, 2013

I spoke at a Euromoney conference on inflation-linked products last year. My thesis at the time that deflation is the real problem and that inflation isn’t going to be a concern – which has largely proved right – was out of step with most people at the conference. I still believe this is the case. And as I prepare to attend this year’s conference on the same topics, I have begun to think again about the deflation and inflation issues. This post on how the US becomes the next Japan is an outgrowth of that thinking.

Let me lay out my fundamental argument here and try to expand on it below.

We live in an endogenous money world. That means that demand for and supply of credit – not base money supply – largely determine broad money supply and credit growth aggregates. In our post-crisis world we are seeing this.
Increases in base money will not spur credit growth. The attempts to spur economic growth via credit growth are at heart based on the money multiplier fallacy that sees credit growth as dependent on base money growth. But in an endogenous money world this hasn’t worked and it won’t work.
The problem is private debt. You cannot get broader credit and money aggregates to grow in a world of restrained credit demand/supply no matter how much QE you do to add to the monetary base. The system is constrained by high levels of private sector debt and the attendant balance sheet issues at financial institutions servicing that debt.
So what happens in the next downturn? I have always believed this is the critical question. As I laid out this thesis mentally, I asked myself what happened when you have near-zero rates, extraordinary levels of monetary ease and liquidity, increasing public debt, and high private debt when recession begins? I think you get deflation, spiking non-performing loans, credit writedowns, insolvent financial institutions, massive private and public sector pension problems and renewed crisis. This is Japan.

First, let me say upfront that a key factor in all of this is the nexus of low nominal GDP growth and pension accounting. I saw an article at CNBC yesterday that pointed out that S&P 500 companies were underfunded by a record $451.7 billion despite the surge in profits and share prices. It was that article and my subsequent thinking about it which in combination with my thinking about the Euromoney conference that helped me put all of these pieces together. I want to say that upfront before I forget because it is an important piece of this puzzle.

Below I will unpack the flow of this thesis for Credit Writedowns Pro subscribers.

We live in a modern credit system

First, there’s the endogenous money bit. I would put it this way: we have a credit system, not a monetary system. That’s because it is the private credit system that is the driver of economic growth. If you are reading this post, then you are probably familiar with my talk about endogenous money because I talk about it ad nauseam. It’s a critical point that differentiates my thinking about the credit or monetary system from the mainstream and it is what leads me to different economic predictions given economic policy.

The basic point of endogenous money is that money is endogenous to the credit system; credit leads the base. This means that base money is created as a response to the creation of money that results from credit transactions in the private sector. You apply for a loan, in this case, a credit card cash advance. The financial institution – let’s call it a credit card company – deems you creditworthy enough for that loan. It has the capital to underwrite that loan and so it does so. The credit card company credits you the money on its electronic ledger so that you now have more money. A deposit has been created without any base money; that comes later.

If the bank is short of reserves, it borrows them in the inter-bank market. If the entire credit system is at its reserve limit given the legal ratio of reserves to deposits, then the central bank eventually adds the necessary reserves to the system via open market operations. Remember, the central bank is an interest-rate targeting reserve monopolist. It is the only entity that can legally supply reserves to the system but it also conducts policy via interest rate targets. Unless the central bank supplies all the reserves the system requires at a given interest rate target, it will not be able to hit that target. The demand and supply for credit will cause banks to desire more reserves than are available, bidding up the price (interest rate) for reserves in the inter-bank market. SO the central bank is forced to add reserves to the system given this price pressure in order to hit the target interest rate. That’s how the system works.

So, it is credit demand and supply that lead here.

Enter quantitative easing

Now, in today’s situation, things got so dire in the financial sector that the central banks decided they needed to act more directly as a bank intermediary and flood the system with liquidity. Much of this liquidity has been in the form of excess reserves, which the central banks have supplied in exchange for government liabilities. Notice, then that the system has a ton of excess reserves then because of this liquidity. And that means that the central bank can easily hit its target interest rate as there is no upward pressure on rates due to increased credit that causes the reserves in the system to hit its upper limit. Financial institutions can grow credit out the ying-yang if they want to before there is any pressure on the target rate because of all the excess reserves.

There are a number of implications here for monetary policy.

Credit easing versus quantitative easing. First, to the degree that central banks use non-government liabilities for QE, they are engaged more in credit easing than quantitative easing. Ben Bernanke stressed this as QE1 got underway. What he was saying is that QE1 was a legitimate lender of last resort operation that the central bank employed in order to enhance liquidity in the inter-bank market that had dried up as banks refused to transact with each other for fear of counterparty bankruptcy. So, QE1 in the US was an operation in which the central bank swapped excess reserves for dodgy private mortgage bonds and quasi-government GSE-backed mortgage paper. Transforming the asset side of the central bank’s balance sheet by warehousing this paper on the central bank’s balance sheet until the paper expires at maturity essentially swaps the default risk from the private financial sector to the public sector. So that’s why this is called credit easing. QE2, the third easing round via forward guidance, and QE3 are not lender of last resort activities because they don’t represent credit easing (see here).
ECB’s credit easing via government bond purchases. Second, to the degree the ECB has bought less highly-rated eurozone government paper to provide liquidity, they are also engaged more in credit easing than quantitative easing. In the eurozone, where the mortgage sector is less securitzed and warehoused more on bank balance sheets, the liquidity problems are in the first instance on the dodgy eurozone government bond side of things. Think of the purchase of Spanish or Italian government paper as akin to the purchase of GSEs or private mortgage-backed securities. In this case, warehousing this periphery government paper on the central bank’s balance sheet until the paper expires at maturity essentially swaps the default risk from the private financial sector to the public sector. Call it monetizing the debt if you will but it is a form of credit easing and a key role for the ECB to take on. Ostensibly relying on this principle of acting as a provider of liquidity is one reason why the ECB has not been willing to take writedowns on its periphery debt.
QE will not spur credit growth. Third, getting back to the main thrust here on endogenous money, the addition of all of the excess reserves from these actions doesn’t have any direct impact on the real economy. It isn’t reducing private debt. So it doesn’t alter the demand for credit. Removing dodgy assets from bank balance sheets and warehousing them at the central bank does actually increase the potential supply of credit by improving the balance sheets of private sector financial institutions but you still have the missing credit demand component to drive the credit creation process that dominates our monetary system. That means QE cannot directly cause the economy to grow except by altering private portfolio preferences or by enhancing what I call rate easing, the forward guidance that signals to markets how accommodative the central bank intends to be in the future.
Interest on excess reserves. Fourth, to the degree that credit does indeed begin to flow and excess reserves are utilized, the central bank has a problem. If credit growth goes gangbusters for whatever reason, causing the economy to risk overheating, then the central bank has no ability to stop this under conventional means. It cannot increase the policy rate to stop the credit flow since there are all these excess reserves in the system. Credit growth would continue unabated even so because there are a mountain of excess reserves to facilitate that growth. Instead the central bank would have to raise the rate at which it pays out interest on excess reserves to the same rate as the new higher policy rate. This would place a strict price limit on those excess reserves consistent with the central bank’s aim of tightening.

In sum, my view on QE then is that it has no direct effect on the real economy because it puts the base money cart before the private credit creation horse. QE works mainly via the private portfolio and interest rate expectations channels. QE can only alter the supply of credit if done in the credit easing fashion that warehouses risky assets on the central bank’s balance sheet. But this doesn’t alter private demand for credit. As such, QE is a bust as economic policy. It is trickle down economics pure and simple because it can only affect the private sector indirectly as asset prices run up due to the private portfolio shifts and deflating expectations for future policy rate hikes.

Think private debt

So this is where the private debt problem enters the picture. Everywhere you look, you see evidence that household debt is still a problem. This is true in the United States. It is true in the United Kingdom. It is true in Canada. It is true in the Netherlands. And it remains the case most everywhere you look in developed economies. What record low interest rates have done, however, is reduce the debt servicing ratios to 30 year lows in the US. But they have not put households on easy street. While household debt to GDP levels have declined, they have not declined nearly as much as debt service costs. And therefore, this presents a problem.

The decline in interest rates has boosted asset prices like housing and caused debt service ratios to tumble. This makes household balance sheets appear much better than they really are by lowering the debt service denominator in the common household debt service to income ratio. At the same time, rising house prices have lifted many households from an underwater mortgage position allowing them to refinance their homes at lower rates. Both of these factors have underpinned a cyclical recovery to the point where US consumer is now rising.

The virtuous cycle of persistently low interest rates, rising housing prices, improving household balance sheets and the concomitant increase in household credit growth could drive cyclical recovery for much longer than some suspect. If we think of GDP growth as being enhanced by a credit accelerator which comes from higher debt to GDP or higher debt to income metrics, then we have the makings of a prolonged if weak cyclical recovery here. This is a major reason I have reversed myself on my prediction that fiscal tightening would lead to recession this year. In my view, the cyclical agents of house price appreciation and credit growth are gathering storm and they are putting us over the top via increased job growth despite the poor wage numbers we have seen. That should be enough to carry us through until the credit accelerator stalls for whatever reason.

But the bottom line here is you need to be thinking about private debt first and foremost, not public debt.

Here’s a chart I posted in 2010 via Annaly Capital Management, a mortgage REIT company.

At that time, I said the data are not good because it demonstrated that there has not been significant household sector deleveraging. But by the time I presented this chart below, there had been a bit more.

If we have continued cyclical recovery, I do not expect more deleveraging. I expect releveraging as the latest numbers on consumer credit suggest. And that brings me to the Japanese question.

What happens in the next cyclical downturn?

If you recall, the Japanese did not have any deflation or depression-like stagnation until the recession and banking crisis of 1997-1998. Two weeks ago I mentioned on Twitter that In 1989, the five largest banks in the world were all Japanese: Dai-Ichi Kangyo, Sumitomo, Fuji, Mitsubishi and Sanwa. In fact, according to a 1989 article from the Chicago Tribune, all ten of the biggest banks were Japanese. Then the bust happened. The Japanese markets tanked and the economy went into recession but no one spoke about a Depression at that time. It wasn’t until the fiscal consolidation in 1996-7 that things fell off a cliff. Marshall Auerback has his post-Keynesian take on Japan’s 1996-7 episode here which I think is very good. My take is here.

What happened then in the context of ongoing struggles and the Kobe earthquake were enormous losses on housing and financial assets that led to huge writedowns and a wave of insolvencies in the financial sector (see here from our archives). The Japanese decided to deal with this problem via bank consolidation – Daichi-Kangyo, Fuji and IBJ became Mizuho, Sumitomo and Sakura became Sumitomo Mitsui, Mitsubishi merged with Bank of Tokyo and Sanwa to form the Bank of Tokyo-Mitsubishi UFJ, etc – instead of resolving the entire system the way the Swedes did. And the US has taken the Japanese example. This crippled the Japanese financial system for years as these banks were undercapitalized and could not lend to stimulate growth. Eventually, however, they recovered enough to be able to help support foundering US and UK institutions like Barclays and Morgan Stanley in the 2008 crisis with capital infusions.

But all this while the back drop was one of disinflation i.e. declining inflation tipping into deflation in the midst of a long period of anemic growth. Monetarists like Ben Bernanke told the Japanese all they needed to do was QE. They should have been quicker than they were. Then none of this would have happened. I got some of this at last year’s Euromoney conference: “one of my fellow panel members told us that the difference between Japan and the US is that Japan didn’t act quickly enough and that the US has done. I believe this is what Ben Bernanke believes as well. To wit, I would agree that Paul Krugman was right when he said in 2010 that we need $8-10 trillion worth of quantitative easing to get the kind of economic impact the Federal Reserve wants. And given what we know about the reaction to QE in the past, $8-10 trillion of QE is a complete non-starter.

“My view is similar to Stephen Roach’s view, namely that QE represents the kind of policy and thinking that got us into this mess. It is unproven and to date, it has not been particularly effective. Quantitative easing doesn’t add net financial assets to the private sector. Nor can QE target specific sectors of the economy outside of housing or short-term municipal funding. I don’t think QE will work. But I believe the Fed and other central banks believe they need to be aggressive because they want to be able to say they did everything they could unlike their counterparts during the Great Depression. As to the debate over whether QE is printing money, it’s irrelevant. The question is whether QE is effective all on its own. And so far, the answer has to be no. After all, even after years of deficit spending and quantitative easing, we are still in an economic crisis five years after it began. That doesn’t sound effective to me.”

So what do I think happens in the next cyclical down turn with this as background? That brings me back to the pension accounting thing. Low nominal GDP and pension funds are a big part of this. In an aging society in which the private sector is highly indebted, gains in growth from the credit accelerator is constrained. Without an expanding population or rising wages, you get low GDP growth, persistently low interest rates to combat this, and weak inflation numbers as demand-fuelled inflation remains subdued. In short, nominal GDP growth stays low and declines. Given the low rate of inflation and the persistently low interest rates, yields on government bonds stay low, reducing income for investors.This is a big problem for pension accounting because pension accounting is predicated on meeting nominal targets for return on investment. And when you are guaranteed to have lower returns on the bond side of your portfolio, you either have to make up a huge shortfall in your pension funds or reach for yield domestically, reach for yield internationally via the carry trade, or take on risk by rebalancing toward riskier and higher return asset classes.

See my write-up on this from last October because the point I make there is that recession is when this becomes most problematic. We already see the pension accounting problem right now despite huge earnings and share increases. In a recession, earnings decline and shares fall, that puts even more pressure on the pensions, causing them to take losses through the balance sheet that feed into the income statement, further reducing earnings. So you have an environment of lower investment returns, operating income, writedowns for pension accounting and rising non-performing loans from the recession. If your financial system is weak – hiding massive losses via accounting subterfuge and regulatory forbearance – you get Japan circa 1996-1998.

At the same time, you also are operating in an environment of still leveraged private balance sheets and limited political policy space due to zero rates and high deficits. As I put it in November, after reading a Jeremy Grantham quarterly, America is on its way to zero growth with heavy consequences. We are talking about lower earnings, lower earnings multiples, higher municipal and financial sector stress and limited political policy space. To me that spells deflation.

This is why I am warning you that the present policy path is not getting us where we want to go. The over-reliance on monetary policy in a zero rate environment not only sucks interest income out of the private sector, it also flattens the yield curve, causing financial institutions to over-rely on under-provisioning and on riskier or non-interest income plays to make their numbers. Given how great the bank accounting gains look now, it would be a shock if that all goes into reverse as I am suggesting it will in another downturn with rates at zero percent. Forget about credit growth in that environment. Think credit contraction, demand contraction and deflation. This is an inherently deflationary Japanese scenario.

P.S. – Posts from the Euromoney conference last year: