Today, the governance issue is heating up drawing room discussions as well as bringing life to coffee tables. Yet, the tug of war between different tiers of the governance levels continues. All trying to grab power for their own side. The clock is yet to strike the hour of decision. And when it strikes, there will be turmoil and unrest amongst the losing party.
However, no one has yet questioned why the policy should be reversed? Is it because it is inefficient? If yes, on what basis? Or, on the other hand, why was it introduced in the first place? Was it for the benefit of the people or the people ruling the people? The irony of the subject is that the idea behind the devolution programme was that the voice of the people would be heard.
Today this voice is being muffled, without any empirical evidence of it being effective or not. A few studies in this regard have been carried out by various organisations, but are with a limited scope. Firstly these studies have looked at the outcome, in a particular province, at an aggregate level and compared over the years, which ignores the study of the districts. Secondly, these studies have merely looked at the outcomes and not at the allocations.
The particular data set, used for this purpose, is the PSLM. This kind of study would face the weakness of lacking benchmark, as the first PSLM was done in 2004-05, which were a few years after decentralisation. Most of the studies conducted about decentralisation take the outcome as a main variable. This variable is extremely problematic for more than one reason. Firstly, the outcome makes it is extremely difficult to predict quality. It is easy to build a school under a tree and classify it as a school.
Therefore, we would never know about the quality of that school. Secondly, outcomes do not necessarily tell us much about the preference structure of the nazims. The key idea of a decentralisation study would be look at how a nazim structures the budget and how it changes over time because of these reforms, if at all. An alternate variable would be better able to capture the desired effect. This would be the district budget allocations.
These district budgets tell us about the spending patterns, particularly of nazims, who have been re-elected overtime. The idea would then be to see what these nazims are doing to be re-elected overtime. Again, it would illustrate whether it is their spending pattern, or it was their political power that got them re-elected. Therefore the struggle between politics and economics continues.
The next important question is that can the study, using district budgets, be conducted at all. In which case is it even possible to study decentralisation? For the sake of simplicity, it is important to define the different types of budgets that actually exist. These include the current budgets (salary and nonsalary) and development budgets.
The hypothesis that needs to be studied is that if the preferences of people are being taken into consideration, then non-salary and development expenditure should be changing significantly and not the salary. However, in order to conduct such a study, the current and development budgets all need to be consistent. The task of making current and development budgets consistent is not only cumbersome, but is also close to impossible for several reasons.
Firstly, they do not follow the same format, not only across districts, but also over years, within the dame district. All districts follow a different pattern, with no common codification. Secondly the development budgets have categories which are inseparable eg roads and electricity in some cases. Then some categories only have a source of funding illustrated and not the break-up. These are labelled as other sources and put as KPP, DERA etc.
In this case, it is impossible to find out exactly what amount went in which sector. The problem particularly would be that if X amount went into education sector in year 1, then in year 2 Y amount went. But in addition some other funds were introduced which are clustered into programs like KPP, so then we will be underestimating the education expenditure in the second year.
Similarly the Citizen Community Board (CCB) expenditure is clumped together, without any separation, in which case it is difficult to find out how much these CCBs spend on each sector. Again, this estimates the expenditure in each sector. Thirdly, sometimes development budgets are not even part of the budgets books and hence, are very difficult to find. Fourthly, some districts publish these budget books in Urdu, while mostly others make them in English and that makes the task of translating and matching even more difficult.
Fifthly and most importantly the development budgets are not coded according to sectors. The ideal situation is to compare percentage changes in salary, nonsalary and development budgets in each sector. This is not possible because there is no common codification between development and current budgets. It can only be done if one would recode everything, but that is also not perfect and has its own loopholes.
Another important problem with this is that of the administration providing this data. Firstly these budgets are difficult to get. Secondly, the attitude of the administration towards the budgets is not very responsible. There are times when these budgets are permanently missing and there is no record. Thirdly, the administration does not have the knowledge needed to handle these budgets. Sometimes, they change codification, which does not match the previous years, even for the current budgets.
This codification doesn't match previous years and also does not match the PIRFA training manual, which they suggest they use for codification. Using the above mention loopholes and problems, some suggestions can be made. Firstly, more projects and studies need to be conducted in order to test the efficiency of the devolutions reforms. Secondly, the key variable needs to be changed to allocation and not outcome. Thirdly, in order to study allocation, the government departments need to create stronger data sections.
Fourthly, the districts need to create co-ordination among each other, so that there is consistency among districts, as well as over years. In event of ignorance of the above-mentioned suggestions, it is evident that all future policy decisions will be ad hoc and without any empirical basis. This will not only retard progress, but will also make application of empirical studies in this area improbable.
(The author is currently a Fulbright scholar and MPhil/PhD student at the George Washington University in Washington DC)