MattAndrews

Elitism, Meritocracy, and Governance in South African Companies

I want to get back to the messiness of governance.

The issue of meritocracy and elitism in good governance

'Meritocracy' is a key element of most 'good governance' models. The idea behind this is that governments, companies and the like should hire those who are most competent to do the job, not those who are most connected. The assumption is that we can measure—objectively—what it means to be 'most competent'. We say these are the ones who have the 'merit' for the job. Hence meritocracy.

Elitism is often presented as the opposite of meritocracy and is something that many 'good governance' advocates reject. These good governance advocates don't like the idea that some people get key jobs because they are elites—connected, wealthy, enjoying power.

Good governance, as many suggest, requires having principles of meritocracy that replace practices of elitism in key appointments, contracts etc. in governments and the private sector. The idea is to have authority in the hands of those who have the merit, which means that authority is not concentrated in the hands of the elite and is in fact located in a distributed group of agents who enjoy the skills and 'merit' to get the job done.

Meritocracy and elitism in South Africa's listed firms

A few years ago (in 2008) I did some research into the background of directors of South Africa's listed firms. These are people who have the kind of positions many would say matter in society (they have power over how resources are allocated, for instance, and authority). You would want to make sure these folks were appointed on merit, so that they did what was required to keep their companies growing and being successful. South Africa scores pretty well on most good governance indicators (given its income levels) and so one would expect that directors would be appointed based on merit.

Some might expect that this means they have qualifications like advanced degrees, in accounting, engineering and management perhaps. And this is what one finds. As the table below shows, 772 of the 873 directors (89 percent) I looked at in a random sample of 97 listed firms had advanced degrees from universities. 562 of these (64% of the sample) had what I would call targeted advanced degrees (in accounting, law, business, and engineering) that one would expect of directors with 'merit'. There is an issue with gender and racial composition in South African boards, so I broke the data down to show how 'meritocratic' directors in different groups were. One can see that more than half of the directors in all 4 sub-groups I looked at (white women through to non-white men) had these targeted degrees. Over 85 percent of the directors in all sub-groups had tertiary education. The lowest percentage was 86 percent (for white men).

	Tertiary qualification	Chartered Accountant	Law	MBA	Engineer	Targeted degrees (CA, Law, MBA, Engineer, Medicine)
Number	774	285	98	121	78	562
Percent of total	89	33	11	14	9	64
White women (ex 31)	27	6	2	4	1	14
White men (ex 560)	484	219	66	66	58	409
Non-white women (ex 80)	77	16	7	16	4	44
Non-white men (ex 212)	188	44	23	35	15	119

So far, so good. Directors of South African firms seem to reflect the kind of 'meritocratic' profile good governance advocates would support. But this is not the full story.

Directors are not just hired because of their qualifications. Other factors matter, which companies communicate in annual reports to shareholders (which boast about the backgrounds of directors). When one looks at these other factors things get messy and one starts to see the social and political complexities of South African society come out. One also starts to see variables that speak to the presence of 'elites'.

• The primary historic South Africa 'elite' variable must relate to race; one sees this squarely in that 68% of directors were white (592 in total)…14 years after apartheid ended.

• Gender variables also reflect 'elitist' tendencies; 87% of directors were men (762 in total).
  • Interestingly, white men had enjoyed more and more appointment over time. 26 % of the directors in 2008 were white men appointed after 2003. (This is greater than the share of non-white male directors appointed in all periods (23%)). 41 % of directors in 2008 were white men appointed after 1998. (This was greater than the share of non-white directors appointed in all periods (32%)).
• Another 'elitist' variable could relate to age; over 90% of directors were over the age of 40.
  • Interestingly, the average appointment age of a director increased from 41 and 46 between 1994 and 2008. Also, the appointment age of white male directors is significantly higher (statistically) than the age at appointment of other groups in most recent periods (post 2003).
• Another 'elitist' variable could relate to access to capital and ownership of the firm. 36% (317) directors came to their positions in this way.
  • Interestingly, 69% of all non-white directors (183) were members of a Black Economic Consortium that held partial ownership in the firm.

There are some variables I captured to reflect on the many times an annual report would speak of a director's past experience when introducing their profile. I was not sure if many years of experience is a mark of merit or of elitism, but the influence of this variable on director selection mattered a great deal.

• 560 of the 873 directors (64%) had big business background (ten years or more working in a listed company, most often the company in which they became a director or a similar company in a similar industry).
• 167 (19%) had experience in the public sector, which involved a mix of political office and public office (being part of the public administration) (100 had direct, publicly advertised political affiliations). Most of these were non-white directors, over 55% of whom had a past public sector career.

So much data. I know. I'm sorry. Sort of. But let me get to the point.

When one looks at directors of firms in South Africa, one sees a group of people who have both the attributes of meritocracy and elitism. And both sets of attributes seem to matter in getting them where they get to. There are very few white male directors who got to be a director without being able to say that they had merit (lots of experience and advanced degrees). Non-white directors may have found an entry way through the capital of an elite-owned Black Economic Empowerment Company, but most of these directors also had tertiary education and advanced degrees and high level experience working in government.

In short, elites are often also the ones who claim points on meritocracy scales. They go to the best schools, get the most targeted degrees, enjoy opportunities to get lots of work experience, etc. And they often define what it means to have merit as well…. Elitism and meritocracy are thus not mutually exclusive but actually form two sides of the same coin.

Pushing meritocracy as part of a good governance agenda is thus not a simple antidote to concentrated power in societies.

I'd love thoughts on this….

Post 2015 MDGs should be shaped around country-specific problems, not global agendas

Yesterday I wrote a short post on the post-2015 MDG goal setting exercise. I got loads of responses online and offline. Just to be clear, I think the exercise of setting global goals raises many questions: about the role of  governments versus international organizations, the idea of 'eminent persons' making high level policy decisions versus normal people doing the implementation, and the gap between high level planning and actual implementation.

One of the reasons I focused on governments identifying their problems (note problems, not goals) and then beginning a process of finding solutions is because this is the best way to mobilize the kind of energy required for implementation. And no matter how much people want to spin the story, the MDGs lacked implementation power in many places (local money, energy, political authority etc. were not provided to push the global goals forward). In my other blog (that supports my book) I argue that problems are the entry point to real change, which also requires experimentation and broad-based ownership (see a link to the blog here http://matthewandrews.typepad.com/the_limits_of_institution/2013/01/looking-like-a-state-the-result-of-limited-institutional-reforms-in-development.html).

One of the responses to my blog yesterday reflected on this issue really well. It is from Natalia Adler at UNICEF and I include the full comment below (emphasis is mine).  Her references to Mozambique reflect our common experience there, and are relevant in many other countries. Thanks, Natalia:

"These types of global goal-setting initiatives   [post 2015 MDG type initiatives] are rarely able to inspire or mobilize concerted action for change, particularly in everyday actions in public management at the country level (never heard government officials saying they have to change course because of the MDGs...).

But the problem is not necessarily related to whether these indicators are globally or nationally agreed upon. I think the same logic applies to national goals as well... The problem is that it's just difficult to translate these outcome-level indicators into a series of operational interventions that will eventually contribute to their achievement. 

As you know, Mozambique is/was developing a national planning system aimed at bringing synergies among all these plans/goals in hopes of making them more strategic, coherent and linked to the budgeting process. But making even well-designed plans and their goals operational is hard work. Plus, politics and power relations influence the extent to which these plans yield concrete results. In addition, the goal-setting agenda carries the assumption of a ‘clean slate’ often neglecting existing arrangements and commitments (to other goals), which cannot be easily brushed aside to give place to new plans/goals, whether they are agreed nationally or globally.  

I personally don't have a problem with the post-2015 discussion identifying a set of 'global' indicators (e.g. child mortality being one of them). What should change are the targets, which, I agree, should be country specific/defined and focused on existing implementation issues and a realist understanding of the pace of development efforts.  I agree with you. The focus on solving problems can trigger change better than goal-setting exercises..

A Post 2015 Menu of Development Indicators, With Fitted Goals?

I'm going to veer off the topic of governance for a short post this morning.

Last week saw a high level meeting in Monrovia to discuss the post 2015 Millennium Development Goals (if that is what they are called). Yesterday the ever-sharp Charles Kenny and some others were active on twitter discussing some of the ideas emerging from the meeting (and in other spaces where post 2015 indicators are being discussed). Charles was questioning the number of proposed goals that not only identify the specific indicator that will hold for all countries but will also note a wildly perfect set of goals--infant mortality to be reduce to zero, for instance.

My immediate response to this is to scratch my head and wonder what all the discussion of country ownership in development really means. Surely countries should be choosing the specific goals they care about themselves. This meas choosing the area and field they want to see particular impact in and choosing the goal in that field.

I understand the international community wants to have international and global goals, but the goals will be met by actions in individual countries and it is vital that these countries are the ones:

• identifying the problems
• accepting baseline measures of the problems
• defining goals
• counting the cost (direct and opportunity) of setting these goals
• looking for solutions
• owning the process of change
• ensuring feedback and learning throughout the process of change

I would propose that the high level meetings focused on post 2015 indicators develop menus of potential goals, with baseline data for as many countries as possible in all of these menu areas. Then let countries choose the areas they want to focus on and what they see as do-able targets, and let the countries drive the discussion about how they will get there.

If ownership matters in development and change, let's allow for ownership through choice.

Economic Governance: Using unemployment rates as an indicator

This morning I suggested that we could use comparative GDP growth stats to shed light on economic governance (how governments are exercising the authority citizens give them to support the goal of economic growth). I noted that GDP growth should be one of a number of indicators of economic governance as citizens are not only concerned about growth.

Unemployment is another potential indicator of economic governance. It reflects on whether government is using its authority to ensure jobs are provided for those needing them. Note I am not suggesting how government uses its authority to do this, just that it has authority and unemployment rates show if this is being used (could be in providing jobs through fiscal expansion or changing regulations to promote small business, etc. etc.).

I used a similar method as the one discussed in the last post to compare unemployment rates in countries. First, I looked at whether unemployment rates were higher between 09 and 11 than they had been in the prior decade (as a measure of whether unemployment had crept above levels considered acceptable). My own country South Africa shows lower unemployment rates (23%) in the 09 to 11 period than the higher rate (27%) between 00 and 09. Second, I looked at whether the unemployment rate was higher or lower than that in comparable income countries. When compared with forty countries around South Africa's income level, the nation turns out to have higher than average rates (about 11% higher).

As with the analysis this morning, one end up with four categories:

• Countries in green (with scores of 3) have unemployment rates in the 2009-2011 period that were below the ten year average before. They also have unemployment rates lower than those in comparable income countries. The analysis does not suggest that this is because of 'good' governance but governments in these countries are bound to face less pressure than others. A critical outcome citizens often look for (jobs) is in good shape, compared with the past and with other countries, indicating that government is using its authority appropriately. Denmark and Norway are here right now.
• Countries in orange (with scores of 2) have unemployment rates that are lower than they were in the prior decade but higher than they are in comparable countries. Governments in these countries face the challenge of exercising authority in a way that enhances the country's economic competitiveness. They probably have the political room to do this given that unemployment rates are down from the past--more people are in jobs than before in relative terms.
• Countries in yellow (with scores of 1) have unemployment rates that are higher than they were in the prior decade but lower than they are in comparable countries. Governments in these countries have the challenge of dealing with a higher percentage of people out of work who will likely perceive that government is not using its authority to create jobs.
• Countries in red (with scores of 0) have higher unemployment rates than they had in the past and that comparable countries are experiencing. Governments in these countries are likely under a great deal of pressure to show that they can use the authority citizens have given them to deal with this problem. It does not suggest 'bad' governance as the causes of unemployment are not always directly attributed to the actions of government. This situation does create a severe economic governance challenge, however, that is common in many countries right now. The USA, Portugal, Ireland and Georgia are in this territory right now.

We should always question unemployment data like this, and there are some countries I am unsure of when I look at the data, but this kind of comparative analysis is useful--I think--in revealing what kind of governance challenges different countries are likely facing. The differences in colors should show how much variation there is in challenges. These varied challenges should be the basis of varied reforms.

 

Economic Governance: Using GDP growth as a governance indicator

I am moving away from debt now, but still on the subject of governance and the question: What would good governance indicators look like?

As readers will know, I take the basic definition of governance as 'the exercise of authority by government on behalf of citizens.'

Digging into the definition, I ask why citizens give governments authority. My answer is: To produce outputs and outcomes that require some kind of collective engagement.

I think these outputs and outcomes are the things we should be looking at as governance indicators. They are not measures of governance, but they are indicators (something observed or calculated that is used to show the presence or state of a condition or trend). The outputs or outcomes tell us if governance arrangements seem on track or off track, and can be used to guide discussions of needed reforms, etc. I resist saying that weak outputs or outcomes tell us that governance is good or bad. These are value-laden words that I find misplaced in any discussion of governance.

I think we need to look at outputs and outcomes in different fields of governmental engagement, and stop pretending we can create single measures across whole-of-government structures. Governance indicators developed at the country level assume that governance challenges are the same in different domains where governments operate. This is an amazingly simplified assumption. Governments operate in the health, education, road safety, economic management, sanitation, energy, and national defense fields to name just a few. In all of these fields the outputs and outcomes governments are authorized to help produce are different, conditions affecting production are different, traditions and rules of the game are different, and indicators should be different too.

In my own thinking, there are five major 'fields' where I think governance indicators should be developed for all countries—assuming that governments have some role to play in producing outputs and outcomes in these fields, in all countries and at all times. These are broad fields (that break into sub-fields) and are focused on (i) Economic management; (ii) Safety and security, (iii) Infrastructure provision; (iv) Social service provision;(v) Societal participation and engagement.

I think economic management is one of the crucial fields where governments engage, with other agents in the international, private, nonprofit etc. sectors to produce key outputs and outcomes citizens care about. These include stable financial sectors, controlled inflation, jobs, etc. GDP growth is one key indicator of whether government is using its authority effectively in this area.

This is a simple indicator that measures growth in economic activity in a country from year-to-year and indicates (shows the state of) governance (the exercise of authority by government on behalf of citizens). Where GDP growth is low it suggests that governments face a challenge in the way they exercise their authority: they need to do something different if they are to exercise it effectively and contribute to producing the outputs or outcomes citizens care about (jobs, etc.).

It is a measure that needs to be contextualized, however, as the authority governments are given is linked to expectations and to contextual opportunities and constraints. I'm playing with two approaches to contextualize these indicators. First, I measure how much a country's GDP growth over the most recent three year period (09-11) differs from the average growth rate evidenced in the previous ten years (which shows what was possible in the context and sets some idea about the expectations citizens might have…what they authorize government to do). Brazil scores about a quarter of a point higher in recent growth stats than the prior decade avrage, for instance. Second, I measure how much a country's GDP growth over the most recent three year period (09-11) differs from the average growth rate in countries with similar per capita GDP levels in the same period (09-11). This shows something about what is possible in the global economy, which could also influence citizens' expectations and gives a comparative benchmark. Brazil scores about three quarters of a point higher in recent growth stats than other comparable countries, for instance.

The map below shows GDP growth data in this manner (taken from the World Bank), and there are three kinds of economic governance situations that emerge:

• The countries in green and scoring 3 are those where recent GDP growth is above both the country's own ten year average and the recent average in comparable income countries. Countries in this category include most of Latin America, Turkey, Mongolia, and India. The data tells me that these governments are probably not facing huge pressure in respect of the way they 'exercise their authority in economic management'—at least in respect of general GDP growth. Governance in these countries is not necessarily 'good' but it is—in the short period under consideration—producing the goods. It has been given authority to play a role in the economic management arena of the country and short term results of the 'exercise of authority' are quite convincing (at least measured in terms of GDP growth).
• Governments in the orange countries ad scoring 2 (like Nepal, Kenya and the Central African Republic) face a bit more of an economic governance challenge. They have higher GDP growth rates now than they recorded on average in the past ten years but they still lag average growth rates in countries with comparable income levels. Citizens are likely feeling that the economy is better managed in these countries (most of which are emerging from some kind of fragility) but they are still relatively low growth contexts compared with others.
• Governments in the yellow countries and scoring 1 (like Australia, Sweden, Mauritius, China, and Mozambique) face a different economic governance challenge. They have GDP growth rates that are higher than other comparative countries but lower than the averages in the prior ten years. Many of these countries are the ones that have been able to withstand the world's post 2008 economic slowdown and this may be the message their governments use to explain the slowdown.
• Governments in the red countries and scoring 0 (a large group including the USA, Canada, South Africa, Angola, Pakistan, and Mali) are facing serious economic governance challenges. Their GDP growth rates are lower than the average of the prior decade and the average in comparable countries. Many of these governments will also argue that the slowdown is a result of the 2008 crisis but they may face some tough questions about why other countries slowed down by less after 2008. Governance in these countries is not necessarily 'bad' but is not—in the short period under consideration—producing the goods. It has been given authority to play a role in the economic management arena of the country and short-term results of the 'exercise of authority' are not convincing (at least measured in terms of GDP growth).

 

I am going to propose that this is one way of looking at governance in the economic management domain. Other variables need to be considered when putting this entire picture together, however, showing if growth is translating into jobs for instance, if prices are under control, and if economic participation is improving. Would love any thoughts you have on this as it is my 'thought exercise'.

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).

Building State Capability at Harvard’s Kennedy School

The Center for International Development at Harvard's Kennedy School has just launched its program on Building State Capability. This is some shameless self promotion of that initiative, but if you are interested in these blog postings it is a good place for you to visit.

http://www.hks.harvard.edu/centers/cid/programs/building_state_capability

In particular, you might be interested in looking at our new paper, Escaping Capability Traps.
http://www.hks.harvard.edu/centers/cid/publications/faculty-working-papers/cid-working-paper-no.-240

One of the arguments we make is related to the discussion I have been having about complex contexts. We propose that the best way of introducing reforms into complex contexts is not to do loads of ex ante studies, but rather to

• enter the context through problems people care about, and then
• progress iteratively with small interventions that flush contextual impediments out, build capability and confidence and political support, and
• allow constant feedback…ultimately fostering the emergence of new hybrids.

In a sense, this is like saying that you need to muddle through the sea of icebergs in your context, ensuring that every step helps you navigate better in future.

Note how different this is to the way institutional reforms are commonly done: They are pre-designed as solutions, pre-planned over years, with limited opportunity for feedback and learning.