The End of Alchemy by Mervyn King – a Critical Book Review

How can we prevent another financial crisis? In the End of Alchemy, Mervyn King presents new ideas to fix finance. The proposals of the former central banker are surprisingly ill-designed.

“Without reform of the financial system, another crisis is certain.” This sounds like a statement from one of the numerous doom and gloom prophets. But no, you find these words in The End of Alchemy: Money, Banking, and the Future of the Global Economy. The author of the book is Mervyn King, known as the governor of the Bank of England who handled the financial crisis of 2007-08.

If a former key decision maker with an intimate knowledge of the financial system warns about an impending crisis, one is advised to listen. This is why we had a close look at The End of Alchemy and decided to write an in-depth book review. Can Mervyn King provide a viable plan to remediate the issues with our financial system?

Insuring liquidity risk

Kings reform proposal centers around a new form of constraint for banks that deals with liquidity risk. He wants to force banks to insure the liquidity risks with central banks when they issue short-term liabilities. The central bank would then determine the “price” of this insurance by setting haircuts for all of the bank’s assets. In turn, the central bank stands ready to provide liquidity by reverting to the bank’s assets as collateral.

Sounds complicated? It is actually quite straightforward once one illustrates the new liquidity risk constraint with an example.

Assume that bank A has assets and liabilities of $100 on its balance sheet. On the asset side, it has 10$ of cash, 50$ of asset A, and 40$ of asset B. And on the liability side it has issued 80$ of short-term debt (in our example, deposits) and 20$ of equity (see figure below).

As just described, the central bank sets haircuts and insures the liquidity risk. Let’s assume that the central bank determines that asset A is not risky, so it applies a low haircut of 10%. Asset B, however, is considered very risky, and the central bank applies a haircut of 70%.

The central bank has now determined the overall amount of short-term liabilities it is willing to insure and the bank is allowed to issue. It is 10$ from the cash reserve, plus 45$ from asset A after accounting for the haircut of 10%, and another 12$ from Asset B after applying a haircut of 70%. In this situation, the bank is only allowed to issue 67$ (10$ + 45$ + 12$) of short term liabilities, that is, deposits.

In our example, the bank issued 80$ of short term liabilities, which exceeds the 67$ limit. The bank is thus forced to raise 13$ of additional cash via issuing new equity, borrow money long-term – or it could alternatively also sell assets.

Reform of central banking

King describes the new role of central banks as pawnbrokers for all seasons (PFAS). It is a different role compared to the traditional central bank role as a lender of last resort (LOLR). The big advantage of having a PFAS compared to an LOLR is that banks always have sufficient access to liquidity from the central bank to cover all short-term liabilities. Furthermore, the conditions of this liquidity insurance are set beforehand. The central bank explicitly guarantees short-term debt in bad times, having set terms that it might seize enough assets to cover losses in good times.

In the end, the intention of the PFAS and the LOLR are the same: Banks and its depositors would no longer have to fear bank runs. It is an objective many proposals to reform the financial system share – for example the 100% Money proposal by Positive Money, Limited Purpose Banking by Lawrence Kottlikoff, or our own call for a systemic solvency rule in The End of Banking.

Still, there is a crucial difference: King’s proposal is designed to preserve private money creation. As such, The End of Alchemy presents a less radical departure from the status quo. So is this seemingly gradual approach promising? We do not think so, and here is why.

Influencing financing costs

The central bank as a pawnbroker for all seasons has to determine haircuts for individual assets. Those haircuts will become tremendously important for each and every one. How so?

The financing costs are of paramount importance for businesses and individuals. Paying 2% or 5% on your mortgage makes a big difference for your overall financial situation. In King’s world, whether you pay 2% or 5% interest will depend to a large degree on the haircut the central banks determine in their role as a pawnbroker for all seasons.

The effect works as follows.

When you are granted a mortgage, the bank has to finance it somehow. The haircut now constrains the amount of short-term liabilities the bank can use to finance the mortgage. If the haircut is 20%, the bank can fund 80% of the mortgage loan with these short-term liabilities. But if the central bank raises the haircut to 50%, the bank can only fund 50% with short-term liabilities.

In this setup, short-term liabilities are also explicitly guaranteed by the central bank. This is an important fact. As a result, the banks' short-term liabilities become as safe and liquid as money itself. The central bank’s guarantee of short-term debt is the reason, why banks would have low costs to tap this source of funding.

The important question is now: How does the bank fund the remaining 20%, or 50% of your mortgage loan? The bank has to borrow long-term or issue equity. As King’s proposal rules out central bank or government guarantees on long-term bank liabilities, the bank would likely have to pay high interest to investors.

Let us assume that the bank’s costs to finance itself with long-term borrowing or equity would be 10%. If the central bank sets haircuts on residential mortgages at 20%, then the bank will charge you 2% interest (to cover the bank’s funding cost of 10% of 20% of your loan). But if the central bank raises the haircut to 50%, you will suddenly have to pay up to 5% on your mortgage loan (to cover the bank's funding costs of ~10% of 50% of your loan).

Central bank’s decisions on haircuts will impact the financing costs of different assets.

This has huge implication from a political economy point of view. How would you feel about the successor of Mervyn King if she essentially doubles the costs of servicing your adjustable rate mortgage?

In the new framework that King proposes, the central banks would make decisions that determine the fate of the entire economy. Centrally set haircuts will determine whether companies or individuals pay 2% or 5% for credit.

Messing with relative prices

You might object and say that already today, central bankers are “the only game in town”. Their reference rate influences all other interest rates in the economy, also the ones house owners have to pay on their mortgage. You are right, but there is a crucial difference.

Today’s central banks just set one reference rate.

Central banks cannot simply target lower interest rates for mortgages – at least they couldn’t before the emergence of unconventional measures such as quantitative easing (QE), which is actually a reason why they are called “unconventional”. So in the old days, when central banks changed the reference rate, they changed it for the entire economy. They did not have the power to influence whether mortgage owners face relatively tighter financing conditions relative to, say, the tourism industry or energy companies.

It is the messing with relative prices that is dangerous. In a world with a pawnbroker for all seasons, central bankers would have to find the right haircuts for all assets in the world – a mission impossible (in particular, in light of the "radical uncertainty" King stresses in his book).

Setting relative prices is a key achievement of a decentralized market economy. King’s proposal would set up a central governmental body that influences all asset prices. The End of Alchemy suggests a centralization of economic coordination power – the devastating effects of such centralization has been vividly shown by socialistic economies in the 20^th century.

Rise and fall of Central Banks

While central banks will never manage to find the right haircuts for all assets, a governance problem arises as well. A pawnbroker for all seasons could potentially manipulate the financing costs across the economy. The central bank could subsidize individual industries or regions by changing the relative haircuts. It could determine the winners and losers of the economy.

For example, environmentalists could ask the central bank to lower haircuts for renewable energy providers, and raise them for nuclear power plant operators. Homeowner association would lobby for lower haircuts on mortgage loans. And if the tourism industry is going through a hard time, they will no longer ask the government for help; they will ask the governor of the central bank for support.

All of these things have nothing to do with monetary policy, that is, with providing a predictable and reliable price level. Irrespective of whether you think green energy should receive more or less public support: This is a question for democratically elected politicians to answer, not for central bankers.

The mandate as a pawnbroker for all seasons would massively expand the power of central banks. Too much power concentration is dangerous. Government will likely abuse this power to support political ideas. The consequence of this could be serious: The End of Alchemy might eventually result in the end of central bank independence.

In his book, King actually notes that unelected central banks should not intervene in the market allocation for credit. He does not realize, however, that his own proposal would inevitably result in such intervention on an unprecedented scale. Consequently, King does not discuss the implication of accumulating ever more power with the central bank.

Fundamental Problem unsolved

Transforming the central bank into a pawnbroker for all seasons would aggravate concerns around central banks having too much power. Furthermore, the fundamental issue with banking guarantees remains.

Whether as a pawnbroker for all seasons or as a lender of last resort, the central bank insures a sizeable part of banks’ liabilities. Lenders to banks have no incentive to constrain the banks from taking excessive risks – that is, building up systemic risk. It is the old and well-known story, and King does not deny the problem of moral hazard that arises from public guarantees.

To mitigate problems with moral hazard, banking regulation would still be required.

King argues that today’s banking regulation is flawed because it is too complex amid the radical uncertainty that dominates our world. We are still on the same page here. However, King then proposes to abandon complex capital requirements and replace them with simple capital requirements.

The academic literature has actually highlighted the necessity of some risk weighting for years (see, for example, Rochet, 1992). Without risk weights, banks that enjoy guarantees by the public have an incentive to hold very risky assets. In the words of Kim and Santomero (1988):

"We demonstrate that the traditional uniform capital ratio regulation is an ineffective way to control the probability of bank insolvency and, thus, to maintain a 'safe and sound' banking system. The primary reason is that it ignores the individual banks' different preference structures and allows 'risky' banks to circumvent the restrictions via financial leverage and/or business risk." (our emphasis)

So why does King call for a simple leverage ratio? To underpin his argument, King cites a study of the Bank of England. This study showed that a simple leverage ratio was better at predicting which banks would fail during the last financial crisis than the more sophisticated risk-weighted measures of capital prescribed by regulators.

King now draws the wrong conclusions. He does not take into account the endogeneity of the regulation’s effectiveness. Before the financial crisis of 2007-08, the banks’ brightest lawyers, accountants, and mathematicians were tasked to minimize risk weights. Risk-weighted capital requirements were poor in predicting bank failure because banks worked so hard to influence them.

In the run-up to the last financial crisis, banks had no reason to manipulate the leverage ratio. This is why the leverage ratio was better at predicting bank failure than the risk weights banks spent so much energy on. There is no reason to believe that a simple leverage ratio could do the job – actually, even the economists from the Bank of England write in their study:

"The evidence that the leverage ratio sometimes performs better at predicting failure does not lead to the conclusion that relying solely on a regulatory leverage ratio would have prevented the crisis: banks had no incentive to optimise leverage by shifting their portfolios towards riskier assets, because leverage was not a regulatory constraint in most countries. Had it been a constraint, banks might have behaved differently." (Bank of England, 2014; our emphasis)

If regulators use a simple leverage ratio, the same lawyers, accountants, and traders would be tasked to manipulate it. There is actually ample historical experience how easily banks circumvented simple capital requirements.

Basel I, introduced in the 1980s, built upon very simple capital requirements that are conceptually loosely related to the leverage ratio that King proposes. What happened? Even the body governing the Basel Capital Accord, the Bank of International Settlement, concluded that Basel I was a failure. The main strategy banks used to circumvent the simple constraints was to move assets off their balance sheet while still keeping the economic risk. It is the well-known problem of regulatory arbitrage.

Regulatory arbitrage was the reason that after Basel I, we saw Basel II and then Basel III.

But however regulators tried to constrain banks, they failed. Banking always swiftly moved into the shadows. Shadow banking migration has been the biggest hurdle to implement effective banking regulation.

The enormous complexity of banking regulation we see today is the result of a struggle between regulators imposing new constraints and bankers innovating their way around – mostly leveraging on the power of information technology.

King ignores the underlying driver of the complexity of banking regulation. With information technologies, financial institutions are quick in circumventing rules and piling up systemic risk. King’s proposal does not tackle any of the fundamental problems with today’s financial system.

Central bankers are no gods

The End of Alchemy is a surprising read. A former central bank governor paints a bleak picture of our current financial system, and points towards many critical issues. We agree on much of this analysis. In the end, however, King’s remedies are ill-designed and dangerous.

Why does a man with such intimate knowledge of finance and rich experience in central banking come up with such a set of proposals? We can only guess that King might lack some critical distance to the institution he served for over 20 years. While he is quick to point out the failure of banking regulators, he remains notably compassionate for central bankers. The financial crisis was mainly a fiasco for the regulators. So for King, the solution is to give more responsibility to central bankers.

But of course, regulations did not fail because regulators were an incompetent or unmotivated bunch of people. Regulations failed because banks could use technology to move assets away from the reach of regulators – basically, they distributed banking on other balance sheets. Keeping a regulatory grip on thousands of financial institutions, all somehow interconnected, is impossible (see the map of shadow banking in 2010 below). Central bankers would be equally powerless in this whack-a-mole game with bankers as regulators have been before.

The flaws in King’s proposal are shared with many gradual reform proposals. They do not question the basic framework, which dates back to the beginning of the 20^th century. Although obvious in everyday live, they do not acknowledge the disruptive power of information technology in finance. In turn, they continue the regulatory race between banking and the public. It is a race the public will always lose. In the digital age, banking has become a dysfunctional public-private partnership.

Mervyn King now fails to address the root cause of the financial crisis of 2007-08. Transforming central banks into pawnbrokers of last resort will only lead to a bunch of new and serious issues without solving any of the old ones. The End of Alchemy will not end any of the problems we face with banking today.

___

Related Articles

• Reflections on banking regulation: Ever expanding regulation will not reinstate good behaviors in the world of finance. To the contrary, more and more rules bear the risk that banking becomes a business without any moral basis at all. (a guest post by John Nugée)

• The small bug that crashed our economic system: Regulators came up with a long list of contributing factors to the financial crisis of 2007-08. But what if all these factors are only symptoms of one underlying, much more fundamental issue?

PS: On funny note, be aware that Larry Summers has some glowing praise for The End of Alchemy: “Mervyn King may well have written the most important book to come out of the financial crisis. Agree or disagree, King’s visionary ideas deserve the attention of everyone from economics students to heads of state.” As we show in our article Investors Beware, Summers praised Marketplace Lending, Larry Summers' track record in predicting the future is far from spotless, to say the least.