Michael Woodford’s 2010 paper “Financial Intermediation and Macroeconomic Analysis”, improves his theory that was articulated in this book, Interest Prices (2003). He based this on the data and experiences from the 2008 financial crisis and changing ideas in monetary economics since his 2003 work.
I have academic respect for Michael Woodford, this review is in no way a commentary on Woodford himself, the sincerity of his efforts, or his intelligent progress he has made with macro economic science.
However, his theories have issues. Yes, at a tactical mathematical level, everything balances and the numbers correspond to his constructs. However, at the strategic level, I question the the constructs and the logic of his assumptions.
You can view Woodford’s paper here in PDF:
Issues with Woodford’s Theory
His work on Financial Intermediation and Macroeconomic Analysis is correct in the sense we need to account for frictions and spreads when looking at the non-bank financing. I would agree with this underlying premise and the data.
Monetary policy based on questionable assumptions – My main criticism is that he advocate of micro engineering monetary policy by the Federal Reserve based on ideas that are not on sound theoretical footing. He advocates perfecting this through econometric models. .
My contention is even if AI were used to perfect the models and account for all variables it would still not get us closer to understanding human action in the market better than the entrepreneurial discovery process.
I think, his rule based approach is superior to discretionary action, but it does not necessarily lead to a high growth long-term equilibrium or optimal capital structure. This is particularly true when Woodford’s approach is coupled with quantitative easing and other tools of the Federal Reserve, outside the rule-based natural rate framework.
Specifically I have issues with:
- His understanding of Wicksell’s natural rate of interest
- Straffa’s critique of multiple rates of interest
- Price stability being synonymous with Money macro equilibrium
- His understanding of equilibrium and money neutrality
- Central Bank rule based (Taylor) action as an optimal equilibrium course
These objections have not been addressed in his 2003 work. Therefore, to add another layer onto this with improve model’s I find it unproductive. However, I will give a summary review below.
What Woodford achieves in his paper on Financial Intermediation
He refines some of the ideas of the equilibrium rate of interest in line with an IS-MP and IS-LM model and adheres to the new Phillips curve, introduces multiple rates of interest, (not in the Straffa) and financial intermediation friction.
Woodford, “Financial Intermediation and Macroeconomic Analysis”, 2010
This kind of model provides a straight forward account of the way in which a central bank’s interest-rate policy affects the level of economic activity and also the inflation rate, once one adjoins a Phillips curve to the model).
Highly aggregated models do not represent human action
However, the assumptions of this model still are in question. The equations workout on a theoretical level, for example, merely altering the slope of the curves, and the precision of the model’s equations are refined. However, this is all on an abstract theoretical level and when markets are in a dynamic and changing environment, data from the past can not necessarily create predictable outcomes for future policy.
Regarding, individual intention and action and future outcomes, economists are notoriously poor prognosticators. If this is the case, policy should not assume so much power.
To bridge this Neo-Keynesian construct to the real world of diverse markets and all the effects and complex human interactions which are seen and unseen are hard to quantify and prove. The theory works in an equation, but in the context of this theory supports an optimal policy decision for long term high growth and real market prices, at the aggregate level it is hard to make a determination.
This ultimately stems from a misunderstanding of what money is and how it functions in real-world markets. This goes beyond improved models for representing financial markets with fiction or without friction.
IS-LM and the Phillips Curve
Further, ideas such as the Phillips curve and the IS-LM model are precise on the theoretical level, but they are just one conceptual framework which is based on a construct for current central bank fiat money. There are criticism’s of both models, for example, criticisms of the IS-LM, however, this is beyond the scope of this post. IS=LM is a diagram for intermediate economics to understand, but to take this into serious theory, the objections and there are many from Gregory Mankiw, mild criticisms to Austrian school major criticisms.
Why banking is important
Firm and entrepreneurial capital financing for the expansion of business come from three sources, retained earnings, bank financing and firms going directly to the market. Retained earnings being the largest component. However, economic analysis and a theory of interest rates is focused on direct and indirect financing because these sources are considered on the margin. Economic decisions are made on the margin.
From these three sources, Woodford focuses on direct external financing over indirect financing, and his most current rendition with the inclusion of friction.
Monetary theory with no banks – However, I question the subsidiary role he places on banking. This seems to support his 2003 work. He claims to revisit Wicksellian theory, but private banks is not a center of his model, only the central bank and direct financial markets. I would have to see more data and evidence to convince me that decisions made at the margin in the credit market are made by non-bank financing. This is a broad stroke of the brush to assert that. Does this apply to all markets and all scales of banking and relevant investment activity or just the activity he looked at or what is good for Central bank policy as it exists today? Does this theory support the current paradigm? What is the US did not bailout the banks and the Federal Reserve did not pump money the way it did, perhaps the banking system would have reorganized into smaller, more efficient banks (Richard Werner’s bank prescriptions), that would support a more positive Gini coefficient. Woodford’s theory is specific to a line of thinking in our current political reality not a universal economic theory of financial markets.
Woodford’s argument is his emphasis is on direct market financing because the growing total dollar usage of direct financing compared to bank financing in the and its perceived role in the 2008 crisis. Therefore his model is constructed with this bias. However, this is not logically rigorous.
Consider how retained earnings is disproportionately large compared to all aspects of external financing, direct and indirect. However, economic focus is largely on external financing. This is because a basic assumption of economics since the marginal revolution as people make decisions on the margin.
That is, what exerts the most influence over markets in terms of movement may or may not be the total dollar amount, but what has the most marginal effect as firms make decisions on the margin.
Woodford does reference studies on both sides but misses this fundamental point. That is before we focus on one construct such as direct financing, economics needs to validate if that is correct. Even if leverage is a factor.
This might be, but it also might be similar to retained earnings in a way which firms finance but the marginal or driving effect of banking and indirect financing should play an important role in the model
We can not assume that based on growth rates of total dollar amounts of funding that this particular method has the most marginal entrepreneurial or decision making an influence. It may be true or it may not be. I think further empirical study at a micro level would be required. What seems to have had the greatest influence can only be known when you see all the influences of human action, rather than raw aggregate data.
The market is continually changing. Indicators have different weights based on changing market conditions. Even if direct financing is growing in numbers, economists need to always weigh marginal analysis before an assumption is made.
This is analogous to any financial ratio used in stock market analysis. As times and economic conditions change in a dynamic economy different variables and ratios are better indicators than others. Before one can make an assumption of where the emphasis should be placed in a model careful study needs to be made particularly at a firm level to understand where the marginal decisions are being made.
However, Woodford concludes and starts the foundation of is model is reasonable.
“Hence, what is needed instead is a framework for macroeconomic analysis in which intermediation plays a crucial role; in which frictions that can impede an efficient supply of credit are allowed for; yet at the same time one which takes account of the fact that the U.S. financial sector is now largely market-based”.Michael Woodford
Financial Intermediation as the center and important part of Interest Rate transmission
This crucial role of financial intermediation is acknowledged. However, where the weight and emphasis is place is what is questioned. How much explanatory value is placed on financial intermediation compared to a bank or non-bank indirect financing when you are looking at it from a marginal perspective.
Woodford argues a new theory:
“the most important marginal suppliers of credit are no longer commercial banks, and in which deposits subject to reserve requirements are no longer the most important marginal source of funding even for commercial banks”.
Bifurcation of Interest and the total cost of borrowing
Woodford builds a new theory of Financial intermediation in the context of IS-MP to include financial frictions.
However, at the core of this theory is an equilibrium interest rate where interests for savers and borrowers are bifurcated to represent friction.
The spread between the two rates can cause counter-intuitive results and affect Fed Fund targeting Policy. Specifically, rates can fall but the cost of borrowing can increase because of the spread increases. Conversely, the Fed funds rate could with the objective of restrictive policy, however, the increase and the cost of borrowing could fall because of the decrease in the spread. This was observed in the pre-crisis and crisis era.
For Woodford, it is the total cost of borrowing, that determines credit expansion. This total cost includes the borrowing rates but also the spread. When making analysis or policy one has to look at the picture to determine, if the supply of intermediation will tend to increase or decrease.
This bifurcation in an IS-MP model of financial intermediation helps give a more accurate model.
Like the Keynesian model of Wage and Price frictions, Woodford has a model of LD and LS frictions when considering the Fed Funds rate.
Financial shocks that are amplified
Woodford suggests a theory of amplified financial shocks of supply of intermediation ultimately affect aggregate demand.
“The dependence of the supply of intermediation on the capital of intermediaries also introduces an important channel through which additional types of disturbances can affect aggregate activity…that shocks that might seem of only modest significance for the aggregate economy… can have substantial aggregate effects if the losses in question happen to be concentrated in highly leveraged intermediaries, who suffer significant reductions in their capital as a result. ” (p.19)
Woodford’s theory articulated is new in details, but overall the concept is not new. He goes onto describe the “vicious spiral… the resulting contraction of aggregate output may result in further losses to the banks, further reducing their capital, and hence tightening credit supply even more” (p.20).
Nothing more than mainline Keynesian idea dressed in aggregate equations
In one sense this is a rebranding of a Keynesian idea, this time supply of credit negative multiplier effect. Financial intermediation is an area that is leverage and have a disproportional effect on the market movement and in a cumulative fashion.
Like Richard Khan’s Keynesian consumption multiplier, Woodford transposes the known Keynesian ripple effect logic to the supply of intermediation.
To some extent, the supply of financial intermediation is analogous to a Keynesian wealth and income effect or some sort of multiplier.
An increase in aggregate economic activity will generally increase the value of intermediaries’ assets (loans are more likely to be repaid, land prices increase with increases in income, and so on) and hence their net worth. This will allow additional borrowing by the intermediaries, and hence a larger volume of credit for any given credit spread. (p,16)
The 2008 Credit Cycle
The Fed Funds rate did not act as a perfect tool or measure because of the spread but also another observed phenomenon.
That is shorter terms rates do not affect investor decisions as much as thought. Rather, ” level of long-term interest rates, which in turn depend on the expected average level of short rates over the coming decade, rather than the current level of short rates alone” (p.20)
This caused borrowing rates to fall with an increase in rates
Woodford understood the Wicksellian framework. He should have know before the 2008 crisis. His theories have empirical evidence, but as you see every crisis is different in a subtle way. It is not that a new exactness would bring the central bank closer to an optimal societal outcome not to mention, productive and allocative efficiency.
- However, his theory does ex-post explain why rates were too low for too long after rate increases. As a Monday morning quarterback or an armchair historian his theory is excellent.
For example, from 2004 to 2007, though policy the Fed Funds rates increased. Yes Woodford makes a good point that frictions or spreads could to be accounted for to have a more accurate model. However, the underlying idea that the Fed Funds target was too low is still the main idea.
The inescapable issue
However, I continue to suggest, regardless of the new tactical improvements in modeling, the strategic picture is credit was too expansive for equilibrium let alone money macro equilibrium.
What if the natural rate was really 6% or 8% at that time? It would make no difference in the larger picture if frictions were not calculated. What would have mattered was the Fed Funds rate was brought up to a proper level to match the natural rate of interest net of all adjustments, which it was not. In retrospect, it was not even close.
In fact, in light of Woodford’s new theory we could call the Econometrically derived natural rate the ‘net natural rate’, if you consider frictions. The strategic picture is still the same. Either, the natural rate is not something that can remotely be measured in any context or model estimates were off and continue to be off.
Therefore, even if you enter in frictions, the larger framework might be some adjustment to the natural rate calculation, but still, my premise is, the Wicksellian natural rate was significantly underestimated.
Woodford’s rebuttal is explained in the context of his model.
“an outward shift of the supply of intermediation schedule XS was responsible, rather than a movement along this schedule in response to a loosening of monetary policy.” (p. 22)
Further, Woodford argues :
“the Fed’s increase in the funds rate over the period between 2004 and 2006 did less to restrain demand than would ordinarily have been expected”
Multiple interest rates
Woodford’s understanding of multiple interest rates is not based on the idea that each entrepreneur might have a specific natural or equilibrium rate, rather there are different rates between investors and savers. This would be a Straffa or even Robert Murphy understanding. Rather multiple rates of interest to Woodford is simply stratification of a singular conceptual policy rate to account for frictions.
“Suppose that instead of directly lending to ultimate borrowers themselves, savers fund intermediaries, who use these funds to lend to (or acquire financial claims on) the ultimate borrowers. Then it is necessary to distinguish between the interest rate
i s (the rate paid to savers) at which intermediaries can fund themselves and the interest rate
i b (the borrowing or loan rate) at which ultimate borrowers can finance additional
Like the gambler who things they can game the system
Which is a key point, that modeling of aggregate values, no matter how precise based on the last crisis will not be the same for the next crisis. This is because entrepreneurial optimizers, profit maximizers will find the next avenue to exploit. These exploitable holes exist because of monetary excess resulting from a departure from free-market banking will create a crack in the structure, and not always in the same place.
The system is design for smart individuals to work around it and take advantage of it. However, the more convoluted it becomes the less just it also becomes. This is why perhaps the focus should be on sound money rather than perfecting the a system that can always be gamed by the next generation of smart profit maximizes.
It is not that economic science should not model, rather we need to a course correction. The models should be based on the acknowledge that the natural rate framework, as seen by econometric models of R-star is not a meaningful tool. To build a model on this will not temper future business cycles or result in a higher growth model, rather a misallocation of resources.
Woodford’s Model for future monetary policy
i) the current value of the “natural rate of interest” – the real interest rate required for output equal to the natural rate, in the absence of financial frictions – converted into an equivalent. One might alternatively define the natural rate as the real rate that would be required for output, nominal interest rate by adding the current expected inflation rate, and
(ii) the current interest-rate spread. (p.24-25)
Woodford’s new policy suggestions include not just interest rate targeting but spread targeting. If policy rates are at zero bounds then Central banks could intervene in market by favoring “the extension of credit to intermediaries by the central bank, on easier terms than are available from private creditors.
…Such a policy can relax the constraint on the size of intermediary balance sheets resulting from limited capital in the intermediary sector, by allowing increased leverage. “(p.27)
I question if the continued path of more complex interventions into market mechanisms is the best policy for long term growth. When this type of policy is known, opportunity can arise where economic agents expect the policy act in a way that would either change the policy action or would not result in desired effects in aggregate or in an optimal free market outcome of allocative and productive efficiency.
Reducing spreads is, in essence, returning a profit margin for firms that are working to provide a market service. It is favoring one group over another for a policy outcome which is questionable in the context of optimal growth.
Woodford acknowledges that it is not a long term policy for central banks to hold non-liquid non-treasury assets, however, the justification being a welfare effect to society if the central bank deems the market is not providing enough credit.
Woodford concludes that until increased regulation and financial supervision is in place, to have the policy tool of spread reduction to enhance rate targeting is recommended.
Zero Bounds Policy
The policy of targeting spreads as a zero bounds policy brings into question if the central bank should have put themselves in a zero bounds situation in the first place.
If a central bank as exhausted its primary monetary tool by putting themselves in a zero bounds position, perhaps it is based on their political motives to jump-start the economy as fast as possible instead of what is good for the long run optimal. This includes not only inflation and output but true allocation based on value. I question if the adverse trend in the Gini coefficient is not correlated to the central bank mis-estimating a fundamental policy guide as the natural rate of interest and whether a rapid and persistent decline in rates to a zero bounds situation did not exacerbate that.
Therefore, a further theory of fine-tuning a policy which is questionable comes into play, better is to examine the natural rate and rate targeting policy.
That is the cost of borrowing increased simply because the natural rate in certain markets was still higher than the fed funds rate increase. Wicksell would maintain an increase or decrease the money rate of interest needs to be seen in the context of the natural rate.
Even if the intricacies of the money market model did play a significant role, it comes back to the point that, R-star an econometric proxy of the natural rate was not high enough to keep the inflation of asset prices in check.
That is an economy with fiat money at an aggregate level will not manifest money macro equilibrium with econometric estimates of the natural rate of interest, no matter how precise the model seem.
The next crisis will exhibit different subtle features and characteristics that will cause this macro disequilibrium to prove out. It is precisely because disequilibrium with rational and forward-looking decision makers that optimize behavior, that investors will circumnavigate counter active policy.
The underlying premise being if there is a disequilibrium in the market for money, it will manifest in a new way. You can suppress the symptoms, but the disequilibrating effects, whether it is seen in, entrepreneurial efforts being steered towards non-market optimal ventures, the Gini coefficient, or asset bubbles will be there will be a crack in the structure. It will result in productive and allocative inefficiency, even if the economy seems to be in equilibrium.
If it did not we would all be able to predict the next crisis and policymakers would be able to steer us clear.