MattAndrews

Debt and Acceptable Debt: Why Some Countries Don’t Respond to High Debt levels

I just posted on my proposed approach to thinking about and measuring governance in relation to national debt. I thought the following figure might be helpful in thinking through the idea. It shows in blue the 2010 level of Debt as a % of GDP in about 70 countries. In red it shows how far the 2010 level departed from the ten year average (where the departure is measured in number of standard deviations). The graph shows that countries like Japan, Greece, Italy and Jamaica had relatively high levels of debt in 2010 (high blue bars) but these numbers may have been considered quite acceptable given debt to GDP levels in the prior decade, which were high/up and down. Governments in these contexts are expected to take time responding to high debt levels. In contrast, Germany, the UK, Ireland and the USA had higher deviations from the norm (the red line) even though the actual level of debt/GDP was not as high as other countries. Governments in these countries can be expected to respond quicker to debt issues than others, because their authority to incur debt is lower.

I guess it's like talking about 'governance on a leash'…. And the leash is just a lot longer in countries like Greece than it is in countries like Ireland… because of past experience, political 'rules of the game', etc.

Going beyond good and bad governance: still using debt as an example

I think the past few blog posts on debt may be evidence of me getting too engaged in one area. Oh well. I think it is super-interesting and important to think of how a government's debt management reflects on its governance. I've had some online and offline responses to my posts that have kept me thinking in this area, and I thought I'd share the thoughts. The risk, as always, is that the fresh ideas make less sense than I'd like. Not a bad risk, however, for those who believe in PDIA: We have a problem (thinking constructively about governance and debt) and my goal is to think about it iteratively to solve it, with (hopefully) lots of feedback and learning.

So, back to debt.

The always-smart and inspirational Natalia Adler reminded me to focus on outcomes when talking about governance, which I think is right (see long-past posts). I think debt levels are an outcome or output indicator. Debt data reflect the way government uses its authority to raise external funds on behalf of citizens. The process of incurring debt is certainly a process, but the final level of debt is a stock that has major implications for many things: How much a country can do now (because of the funds raised), how much the country must pay now, what must be paid in future.

Philipp Krause reminded me that debt levels are not a simple measure of governance. It is hard to say that high levels of debt are good (or great, as Philipp suggests) or bad. It is contextually relative.

Nadim Mata noted that saying 'governance is relative' is a dangerous thing. My word last week was 'uncomfortable'.

So: where do we go with the idea of governance…focusing on outputs or outcomes that people care about yet in a way that accounts for contextual factors…yet is not saying 'all is relative'?

Here is my idea. Love to know your thoughts.

First, I ground my reasoning in theory: Governance is about the exercise of authority by government on behalf of citizens. Drawing on Dani Kaufmann's work, it is about the way governments work within the context of established authorizing mechanisms and rules of the game. In respect of debt, I argue that governments have authority to raise money to further the interests of citizens. They are given this authority because it is clear that there are situations where debt is needed. The authority is not limitless, however, as there is a downside to too much debt (its cost). Different contexts are likely to have different thresholds of what is acceptable in terms of levels of debt a country can incur. This is determined by informal cultural variables, norms, practical constraints, formal rules that might restrain debt, etc. When we look at debt governance, therefore, we look at the way governments exercise the authority to raise debt within the rules of the game in a given context.

Second, I looked for an indicator: I wanted the indicator to be easy to understand, for starters, and contextualized as well. The standardized content comes in the form of the level of debt (as % of GDP). This is the number most people talk about when reflecting on debt issues, which makes it an easy starting point. The conventional approach is to show that high debt = bad and low debt = good (at least this is how I see the conventional position…). But we know it is not so simple, because some countries (like Japan) have had really high debt to GDP ratios for a long time and seem to continue amassing debt. The Japanese experience shows that some groups of citizens authorize their governments to run up high levels of debt. They do so because of various other characteristics (including the fact that Japan is also one of the world's largest creditors…). So, the amount of debt a government can run up before becoming 'bad' is relative to its contextual characteristics and how much and what type of authority it is given.

Third, I asked how we could capture the contextual authority governments have. I started working out some political and social characteristics of different countries and working out the patterns associated with these characteristics. Countries with natural resources like oil seem to have lower debt levels, for instance, and countries that introduced fiscal rules because of major fiscal meltdowns also have lower debt (Sweden is an example). These stories make sense, and for some countries I found myself developing useful categories that helped locate the kind of debt governance one might expect and how this would equate with debt levels (oil independent like Azerbaijan = limited debt; fiscally restrained like Sweden = limited debt; constrained debt like many developing countries who cannot raise debt = limited debt; net creditor like Japan = higher debt). But one only has simple stories with some countries when looking at these kinds of variables, and I found a number of countries really hard to make sense of. Partly, I believe, because the contextual characteristics influencing governance arrangements are imperfectly apparent…like an iceberg we only see some factors.

So, I asked what evidence there may be about how rules of the game manifest in different places. And I thought, simply, to look at what seemed to be 'acceptable' levels of debt in different nations, given actual data from the prior ten years. A country like Japan has averaged over 100% of debt to GDP in this period, for instance, with relatively little standard deviation from this mean. Switzerland has averaged about 22% and there is very little variation between years. What I take from this is that Japanese authorities are authorized to incur high debt levels but Swiss authorities are not. Ireland had an average of about 45% but the variation between years was really high, indicating a major decrease in debt/GDP levels from 61% in 1998 to the high twenties in 2007. What I take from this is that Irish authorities were authorized to decrease debt to GDP ratios in the 2000s, after experiencing high levels in the 1990s. A return to high levels would not be well received, given that this was beyond what was now being authorized.

I developed a variable out of this analysis that shows how much the 2010 debt levels deviate from the level that seems 'accepted' in the past decade. I used a similar basic formula to that employed in a 2010 Oxford Development Studies article, which is essentially the z score (2010 debt = average debt)/std. deviation. I know that I am assuming a range of things about distributions etc. that create problems but I'm forcing myself to simplify so for the time being let's forget this. Japan scores 1.51, suggesting that the level of debt is 1.5 standard deviations above the 98-08 average. Ireland is 2.35, indicating that its 2010 debt levels were nearly 2.5 standard deviations above the 98-08 average. Note that Japan's debt levels were about twice those of Ireland at the time (174% of GDP as compared with 8% of GDP). Theoretically, however, it would appear that Ireland's authorities were way above their authorized range…even at a lower level of debt…and one would have expected some kind of change.

Consider the following figure that shows 2010 debt to GDP levels and deviations from 'acceptable' levels in about 70 countries.

I like this way of thinking about governance in respect of debt management. It allows us to look at an output or outcome of government activity (the debt it has incurred) but in a contextually specific and relative manner (examining if the government is behaving within the scope of authority it enjoys, as reflected in revealed performance over the past decade):

• It might help explain why the Ireland has responded to its smaller debt burden quicker than Greece has responded to its large debt burden (the Greek debt is more accepted, given the history of high debt and high variation).
• It could also explain why debt levels are a key issue in countries like the USA and UK even though many economists say both countries have 'technically' got space to absorb more debt. Given the recent history, citizens in both countries have become used to having lower levels of debt than they had in, for instance, the 1980s. This means that governments in these contexts are operating in contexts where higher debt to GDP levels are not going to be easily authorized and even a debt to GDP level of 88% draws attention in the USA. At this level government is operating about 2.5 standard deviations above its average debt level in the past decade, stretching its authority to raise debt.
• The analysis also allows us to look at countries that use debt sparingly, and we can better understand the stories here. A country like Sweden has debt to GDP levels of about 40% but this was about 1.5 standard deviations below the 98-08 average. This reflects the fact that Sweden has become a more restrained government in this period, where authority to spend and incur debt has been systematically reduced. Cyprus, which I believe is about to default, is also in this area below the line because of higher debt ratios in prior years.

There are many questions I have about the details in the figure above, but what I like is that it allows us to tell a story about how governments are performing in light of the revealed authority in each context. One might use this to say that governance is more risky as it moves further away from the 0 point on the vertical axis; governments aware of this risk will be expected to either reduce their debt or begin serious political communication about the need for more authority to incur debt. This is, I think, what you see now in the USA. It's not that governance is good or bad in this context: Just that authority is being exercised at an extreme…

Uncomfortable thoughts on governance (Can we ever say it is good or bad?)

My brief few posts on debt and governance have helped me illustrate (to me, at least) that it is really hard to say that governance is good or bad.

As I understand it, governance is about the way governments exercise authority given to them by citizens.

I looked at debt because governments are authorized to raise debt on behalf of citizens. My first blog post on the topic noted that developed countries have higher debt/gdp ratios than developing countries. I implied that this seemed like a bad thing. Mostly because thirty years ago high debt burdens were a major reason developing countries landed in trouble, were termed 'bad governance' examples, and ended up beholden to multilateral and bilateral lenders.

But is debt always bad, and are countries that run up higher debt burdens than others badly governed. The ever-smart Philipp Krause thinks not, as reflected in today's Beyond Budgets blog post (What Does Debt Mean for Governance?). Philipp writes, "Matt wonders whether this worsening of rich countries’ debt position says something about their governance. If debt/GDP was a governance indicator, what would it tell us? Matt seems to lean towards thinking it’s not a good thing. I think it could just possibly be a great thing." (the emphasis here is mine).

Philipp's post is worth a read (as is the Beyond Budgets blog) and I think you will see some meeting of minds in his argument and my posts from yesterday and the day before. I would argue quite emphatically that we cannot say that higher debt means bad governance. Or that low debt means good governance.

What these indicators mean for a specific country depends on the country, its pressures, opportunities, etc. And most importantly, what the citizens authorize the government to do. If governments are on a short chain because citizens remember financial calamity with fear then high debt is probably an indicator of bad governance. If the country is coming out of crisis and there is demand for government to lead economic recovery and the government can raise debt cheaply, then surely high debt is not bad governance. (I can't bring myself to say, with Philipp, that debt is great...).

The contextual relativity of this discussion might bother some people who are looking for indicators that are certain and specific, and can imply judgment in a single figure... But I think that governance is a relative and contextual thing, and that any governance discussion needs to face up to uncomfortable  thoughts. Let me suggest, for instance, that we can rarely say that many of our staples of good governance are good or bad:

• If meritocratic hiring is always a good thing, why do we see most governments across the world reserving posts for political hires (and why do we see meritocratic hiring reforms conditional on other things...see Christian Schuster's interesting paper Civil Service Reforms are Thorny). Could it be that the right political appointee is the right civil servant in some positions? and is 'merit' such an objective concept after all?
• If transparency is always a good thing, why do many of the world's better governments value secrecy in some areas? Why do the World Bank and IMF keep negotiations secret until they are concluded in many instances? Could it be that transparency and opacity are just tools or conditions under which different things happen? Could it be that some high transparency countries are just being transparent about stuff that doesn't matter? So the act of being transparent is neither good nor bad.
• Does a business friendly government always mean a better government? many current indicators say it does. But sometimes we want government to be a regulator of business and sometimes a facilitator. So, regulations are neither good nor bad either.

I think we could make a list like this for just about every indicator of governance, and argue that labels like 'good' and 'bad' are oversimplified, misplaced and potentially damaging to the overall narrative on governance.

High debt is neither good nor bad. High levels of merit-based hiring is also not good or bad. High levels of transparency may not be good or bad either. 

Uncomfortable thought about governance for the day: The idea of good governance may not make sense.

Dos the USA exhibit ‘bad’ debt governance?

In my last post I showed that the recent record of debt and debt servicing looks different when one looks at different countries. This makes it difficult to identify 'good' or 'bad' governance in any comparative way. I would rather say we should say that some countries tend towards a more conservative governance in their debt management than others. Other countries have more of a risk-accepting governance model.

The key issue is "what has the government been authorized to do" and, consequently, "when are they doing this well—so that they exercise the authority effectively—and when are they doing this poorly—exercising the authority in too conservative or too risky a fashion."

Let's think about this in relation to the United States:

• Many on the conservative side of US politics complain that the federal government has exercised its authority to raise funds through debt poorly—being too risky. One expects that these commentators want the USA to be in the more conservative domain of the graph below (which I showed a few days ago)…where Sweden and New Zealand currently are. They often say that the US is going to become like Greece.
• Less conservative commentators think that the USA can probably still raise more debt without increasing its debt service costs too much or becoming like Greece. They reference the fact that debt has grown in recent years without growing debt service costs (see second graph, shown this morning) and argue that too conservative a stance could stymie the country's efforts to get out of the economic doldrums.

The political debate about US debt is an important one in that it will determine whether the US continues shifting right on the horizonatal axis (increasing debt) or turning a 180 and decreasing debt. As I have suggested, there are normative angles to this debate…as with any debate on governance…where the question is "what kind of authority does government have and is it running up against limits to authority?"

Interestingly, this is not a purely ideological debate, and there are various factors one can look at to see if the US debt governance currently qualifies as 'good' or 'bad' in a fairly objective sense. Here are two of them:

First, when one compares the US situation today with that of the past, it is obvious that the territory currently being traversed is not new. The US experienced high debt levels between 1945 and 1956, for instance, and saw debt rise substantially in other periods (between 1980 and 1990, for instance). These instances tended to coincide with the overexpenditure associated with wars and with expenditure and revenue implicatos of economic downturns. It seems these factors are acepted as reasonable drivers of acceptable public sector debt. So, the argument in 2012 is that the federal government has amounted debt because of wars in Iraq and Afghanistan and the 2008 crash. Given such events, the current debt does not reflect a government acting outside of its authority.

Second, the US can still access cheap debt to finance spending that many seem to believe is warranted. The US federal government borrows at historically low rates right now and it appears that the public demands federal engagement more than conservative pundits would like to believe. Polls show that citizens are more interested in the government helping create jobs, infrstructure etc. than bringing down the debt (given that the latter is always a lower-rated issue than the former). So it seems that government is being authorized to incur more debt and when it does so it is not necessarily driving up costs (at least not for now).

There are obviously many other points to consider, but if these two make sense then one cannot say that the current governance of debt in the USA is 'bad'. Risky, yes. But not 'bad'. And not worthy of all the hype it attracts.

 

Governance and Debt in the USA, Greece, Ireland and Sweden: 4 contexts, 4 stories or 4 versions of the same story

I blogged about debt and governance in the last few posts.

The core idea behind these posts is to see how we might conceptualize and measure governance in the field of government debt management. Conceptually, the idea is simple: Citizens give governments authority to use debt to further the goals ad needs in running the state. In terms of measures, I suggested we look at the amount of debt (relative to GDP) and the cost of servicing the debt (as the opportunity cost of not doing other things).

I came out with a two dimensional view of the situation and potential scenarios: From conservative governance to managed risk to higher risk governance scenarios.

I suggested that we cannot simply refer to 'good governance' or 'bad governance' in these scenarios but we should use the analysis to ask questions about governance challenges facing different countries.

Some folks asked me offline how I would use the data in this manner, and I thought I'd give some perspectives on this. My first approach is to ask why different countries are located in different places in the debt governance realm and what this tells me about the kind of authority they have (a key idea in my approach to governance) and what the recent and even past trajectory suggests about current status.

Let me start by showing some recent history for four countries, which shows (me at least) how darn complex any work on governance is.

First, the USA. The graph below shows that the USA has consistently spent just over11% of its revenues servicing debt, since 2001.What has changed is the level of debt, reflected in a rightward shift throughout the period. This suggests that the government is authorized to increase debt and can do so without incurring additional costs in the immediate period. The question of course, is how much more debt can be incurred and if the costs will stay at this level. This is an important question and is not just technical but also reflects the political dynamics within the country. It does not suggest that austerity is required, but that the country is in risky territory and authorities need to ensure their mandate is secure as they move ahead. I will get into this a bit more in the coming days, trying to understand why the US government has increased debt in the past decade and if this suggests problems with debt governance or not.

Now, Greece. This story is totally different. Since 1995 Greek debt has been above 100% and debt servicing costs have been historically higher (as a percentage of revenues) than in recent years. The government reduced servicing costs in the period between 95 and 99 and managed to keep debt at around 110%. Then, between 2000 and 2007 they carried on decreasing servicing costs but saw debt/gdp ratios rise to about 125%. Sine 2008 both debt/gdp ratios and servicing costs have been rising, which reflects the real trouble this country is in. The governance challenge is two-fold: bring the debt down and convince the public to endure higher servicing requirements. Both demand austerity.

Now, Ireland. I have not blogged about Ireland to date but it is an interesting case and one that several of my past and present students are interested in. The situation here is again completely different to the other two. Ireland decreased its debt and debt servicing burdens substantially between 98 and 07, partly because a growing economy influenced the measures but also because there was pressure for the government to show itself as fiscally disciplined. This was part of its story in the past fifteen years of engagement in the EU. One can see that the government was locked into appearing (at least) like a conservative governed debt context (a Swedish story, perhaps). Between 08 and 10 one sees that debt and debt servicing levels have risen substantially, largely because of problems in the economy. The debt governance challenge has changed into one of containing rising debt and servicing costs in the context of a slow growth economy. The question is whether the government can change its approach to respond to this challenge. (I hear it is….)

Now, Sweden. I have to cover Sweden because it has been a focal point of so many ideas about what the USA should do to escape crisis and get its house back into order. The graph below shows the situation is again totally different. I only show data from 1995 but you can see that Sweden has been on one consistent path towards lower debt and debt servicing levels since that time. You can also see that it has been in a different part of the graph to the USA or that entire period as well The USA government has always had higher debt servicing costs (at least measured in this way) which seem appropriate in the context. The Swedish debt governance challenge seems to be consistent since the early 90s when crisis was deep and prolonged: Get debt down and keep it down.

What is the point of these graphs and analyses? First, to show again that politicians who say the USA risks becoming like Greece should stop making such commentary. And commentators who say that the Swedes have lessons for the USA to learn from should think carefully about why the two countries seem to have such different looking legacies: what does this say about what is appropriate in the different contexts and how might this impact what kinds of governance reforms are possible. It seems to me that governance stories are highly contextual. The four stories of debt and debt governance I see here look like different stories that demand different interpretations and different responses. They are not 4 versions of the same story that can be interpreted through the same lens and lends itself to the same solutions.

Debt and Governance: The USA is not Greece; but it isn’t Sweden either

I blogged about debt and governance a few days ago, suggesting in a highly simplistic fashion that we should look at government debt levels as an indicator of governance. The argument is simple: Governments are authorized by citizens to raise revenue through debt when it makes sense and to manage the debt exposure to ensure the country's fiscal dependence.

I reflected on data that showed developing countries have lower debt to GDP levels than many developed countries do. This is a reversal of the past, where developing countries typically found themselves in debt problems that were arguably construed as reflecting a failure of governance.

Now, what do we make of high debt figures in developed countries? Is it also a failure of governance? Philipp Krause responded to my simple post by saying that it may be the exact opposite. If governments like Japan and Italy can borrow at cheap rates to levels of 150 - 200 percent of GDP, surely it shows that they are seriously creditworthy and the government is just getting better leverage in financial markets.

Certainly an interesting argument. Of course, there is a cost of borrowing, and every cent one pays in servicing debt is money that could be spent differently. Some countries manage this cost better than others, by structuring their debt appropriately, etc. Even when they do this, of course, many argue that governments are just deferring the debt costs to future generations. And even well managed debt portfolios can become costly in the short run as well.

This is important because today's governments are not elected by future generations, and governance is most simply about exercising authority on behalf of those who gave it to you... today's electorate. Unless, of course, today's electorate indicate a concern for the future.

In reflecting on this, I'm going to suggest that we look at two dimensions of 'debt and governance' in today's terms when thinking about this issue.

• The first dimension is the amount of debt a government has incurred, as a percentage of GDP, relative to peers.
• The second dimension is the cost of that debt in today's terms, measured as the percentage of revenues collected, also relative to peers.

The combination of these dimensions is shown, for about thirty of the world's wealthiest nations, in the figure below. (All data from the World Bank's World Development Indicators).

The first thing I see in this figure is that there is a positive relationship between the size of a country's debt relative to GDP and its servicing costs—at least in this one period cross section. In other words: the bigger a country's debt today, the more a country tends to pay today. So, a country like the USA pays about 11 cents of every dollar raised revenue on interest to service its debt—compared with Sweden where less than 3 ӧren are spent n interest, out of every Krona collected as revenue.

I like the table as a way of indicating the governance challenges of a given country—note I am not saying that the indicator should be construed as reflecting good or bad governance. There are sixteen quadrants in the way I show the countries and a country's location says different things about its debt governance:

• Countries in the bottom quadrant are experiencing debt levels and debt service costs that are lower than the average for wealthy OECD countries. At a stretch one might say the debt here is being conservatively governed (which conservatives might call good governance). Sweden and New Zealand are in this quadrant in 2010.
• As one moves up and to the right you are looking at countries with higher than average levels of debt and costs of debt servicing (as percentage of revenues). The United States fits in this area, where its debt levels are abut twenty percent higher than the average in wealthy countries (measured as a percentage of GDP, of course). Its debt service costs are just over twice the average in wealthier countries (at over 11.2% of revenues collected). The countries around the United States are in a riskier part of the graph than Sweden and New Zealand in 2010. Sweden was in this area in 1996—when it was coming out of its 1992 financial crisis. This space might be called a managed risk governance situation where government is extending its authority to borrow and taking the short term hit of extra payments because it can access funds that it believes are needed. In Sweden the resources were needed to get out of a financial crisis that it has now moved past. In the USA the authorities argue that the higher debt is also needed to get out of crisis. The question is whether this managed risk situation will pan out.
• As one moves further right and up countries have higher debt and debt service costs. The inhabitants of these quadrants are few. Iceland, Italy, Greece, and Japan. These countries are in a high risk governance situation where the burden of debt is high and the current cost of debt is also high. Countries in this situation may be able to access additional debt (for now) because of their peculiar economic circumstances (Japan is still one of the world's biggest creditors, for instance, and much of its debt is domestic) but they may also have gone into a debt crisis situation where they cannot access conventional debt (Greece). Risky territory, that may be accepted by citizens but is still costly—now.

For those who like to criticize the USA debt situation and suggest that Armageddon is days away, the graph should show something simple: THE USA IS NOT GREECE. It should also show that the UA is not Sweden, or any other conservatively governed debt regime. The question is whether this is a problem or not, and if governance strategies should change as a result of the USA's position.

For those who are interested in governance, this final question is of utmost importance. This is because it is the kind of product that we should be developing from governance indicators. Strategic questions to guide comparative policymaking, not singular measures that suggest whether a country exhibits 'good' or 'bad' governance. The contextual nature of governance prohibits such simple measures and we should stop pretending these kinds of measures are useful.

Government Debt and Good Governance: Rich Countries Now Lag Poorer Countries....

Regular visitors to this blog know that I spend a fair amount of time thinking about governance and governance indicators. I have proposed a range of ideas for measuring governance in the past. Let me try a new one today.

As in the past, I like to define governance as being about the way governments exercise authority on behalf of citizens. In this respect I think that governments are given authority to raise revenues through taxes and debt (and other means) to responsibly finance interventions in society.

It makes sense then to say that good governance must be reflected in the degree to which governments exercise this resource mobilization and management authority.

A potential measure of this--which is clear to understand--is the level of debt to GDP in a given country. This was certainly used as a measure to argue that African and other developing country governments could not govern in the 'stabilization' years of the 1980s. Many many governance reforms came from such logic.

So I looked at World Bank data on this for about 70 countries and generated the following simple excel graphs. They show how countries with lower GDP per capita tended to have higher debt/gdp ratios as recently as 2001 (with a trend line showing slight negative relationship between the variables). But if you look at the other two graphs you will see that the trend line changed over the 2000s and there is now a positive relationship: Wealthier countries now have higher debt to GDP ratios on average than poor countries.

As with all indicators we should contextualize the measure and i need to say up front that this is the product of an hour or so tinkering around...so don't go presenting this to your class as gospel! This is all the more important given that debt is now the political focal point in countries like the USA. So it behooves us to remember that 'ggod governance' could lead to higher debt to GDP ratios in some times than in others (like after a financial crisis).  It is also important to look at actual data and remember that US government debt to GDP is only about 75 % now. IT IS NOT GREECE (sitting at 1bout 135%, behind Japan at over 150% according to the World Bank). In this list of about seventy the USA ratios are lower than a variety of countries: Hungary,Malta, the United Kingdom, France,Ireland,      Belgium, Portugal, Barbados, Singapore, Iceland, Italy, Greece, and Japan.

I will bog a little more on the US situation in coming days, thinking about the kinds of processes that might help improve the governance of resource mobilization and management when this crisis is over...

In the meantime I really wanted to revel (as a development person) in the fact that poorer developing countries seem to have managed to keep their debt to gdp ratios in check over the past decade. Of course this reflects the fact that the 2008 crisis hit wealthier countries disproportionately. It could also suggest that developing country governments don't try to spend more in downturn because they realise fiscal policy is blunt if you can't actually implement the spending plans you propose (because of some of the process problems I have recently discussed).

The Limits to Public Financial Management Reform in Development

My book is due out in two weeks. As prior blog posts noted, it discusses the limits of institutional reforms in development. 

via matthewandrews.typepad.com

Mapping Public Financial Management Process Quality

One of the goals of next week's Harvard executive course in public financial management (PFM) is to use data in thinking about where PFM reforms are working and where they are not working.

Data always has constraints, however, and we should be careful when using data. We should be especially careful in over-interpreting data, or in drawing too much from comparisons between different contexts, etc. This is especially the case when we look at anything to do with governance or public management, where the same 'script' is used to collect comparative data from different countries. Lower comparative scores may mean genuinely weaker performance or just a poorly chosen or fitted 'script'.

Acknowledging this, we must also recognize that there are increasing data about governance and public management systems across the world. These data give us an interesting window onto the administrative mechanisms in place in governments. This window is helpful when thinking about which countries can learn from which countries, where reforms are working and where they are not working, and so forth.

One such data source comes from the multi-donor group PEFA (the Public Expenditure and Financial Accountability group).

A PEFA Public Financial Management (PFM) Performance Framework was developed in the past decade to assess how developing countries were performing on what the donors called 'good international practice'. Over 80 countries have undergone at least one assessment thatis in the public domain (see at www.pefa.org). 

I have been working with this data for a few years now, trying to work out what it tells us about PFM systems in developing countries, progress with reforms, and so forth. In a variety of papers on the topic, I and others (including those in the PEFA secretariat) have focused attention on 64 scores the PEFA assessments give for the PFM processes in place in a country. These processes range from those related to strategic planning, budget preparation, tax administration, cash management, procurement, internal audit, accounting and reporting and external audit and legislative engagement.

The PEFA scores range from 1(a D) to 4(an A). The following map shows averages of these 64 process scores for about 80 countries where PEFA assessments are public (using means, which are not strictly the most appropriate measure given the ordinal nature of these data). The objective of this is to show, simply, what picture these data give (in a very simplified manner) of which countries have higher PFM process quality (in green), and which have lower PFM process quality (in red) and which are in-between (yellow and orange) given the PEFA analysis.

You should be able to click on the map and play around a bit. Anyone see any patterns worth mentioning? I would love any comments or perspectives.

Public Financial Management Executive Education Course

Next week we have 60 to 70 participants coming to Harvard; from over 30 countries to discuss Public Financial Management (PFM) Reforms. The executive education program is designed to provide new ways of thinking about doing reforms in this complex area. I describe the intention of the course in this video.

I am looking forward to a week where we use data to reflect on where reforms are working well, and where reforms are limited, and how we can move past the limits to more effective reforms in more places.