Pakistan has a long history of launching – but not implementing – policies and plans of all kinds. In fact, if there is one thing we can be sure of when developing a policy (or plan), it must be there is often no plan to actually implement it.
Take any policy or planning document or a five-year plan this country has produced – we have produced many since the 1960s when this process began – and one can easily see this to be the case. Every plan begins with an appraisal of how the previous plan did, and while an effort is made to demonstrate that the previous plan worked (or not), it is often easy to spot that the plan did not achieve even half of what it was supposed to achieve and whatever was achieved was, perhaps, in spite of the plan and not because of it.
It is often repeated in our development discourse that Pakistan designed some masterful five-year plans in the 1960s that were copied by the South Koreans and a claim is made that had Pakistan implemented what it planned, we would have been where South Korea is today. I don't know to what extent the former statement is correct, but I can assume that there may be some truth to the latter.
If one studies the South Korean development model from the late 1950s and 1960s onwards, one is surprised at how flawless and masterful the South Koreans have been in identifying the right policy levers, picking the right winners, and implementing whatever they have planned. And this has resulted in South Korea becoming one of the most technologically advanced countries in the world, with exports – what Korea sells to the world, not what it produces – being more than three to four times Pakistan's Gross Domestic Product (GDP). In the 1950s, fresh after the Korean War, these numbers were probably reversed.
There are many reasons for this failure of policy and planning, and I feel they mostly lie in our inability to understand what a policy actually is. In Pakistan, policies are usually big, thick documents that have a lot of words and hardly any substance – and (almost) certainly no will or wherewithal to implement. In effect, creating and launching policies is, at best, meant to be an exercise in ticking a box and hence an exercise in futility, and at worst, designed to procure and spend resources that one will never be held accountable for. No minister or ministry in Pakistan's 75-year-old history has ever been held accountable for his/her performance or failure to implement a stated policy.
There are several issues at hand here, and I will address these one by one using Pakistan's recent attempt at a National Science, Technology and Innovation Policy (NSTIP) as an example. The NSTIP is a 145-page document produced by the Pakistan Council for Science and Technology (PCST) and issued by the Ministry of Science and Technology in January 2022. I particularly like the fact a National Policy document begins with the following thought-provoking and scientifically inspired verse from Allama Muhammad Iqbal, our national poet-philosopher:
یہ کائنات ابھی نہ تمام ہے شاید
کہ آر ہی ہے دمادم صداۓ کن فیکوں
The NSTIP document was produced by a committee of 17 people – mostly drawn from government and academia – with minimal private sector representation. It is a comprehensive – often verbose and overly prescriptive – document with analysis, policy, and plans all rolled into one. However, the one thing that it seeks to do that is quite novel in the Pakistani context is that it identifies several "policy statements" presented as key decisions that this "policy" seeks to affect. This is a welcome departure from the past practice where any exercise in policymaking quickly turned into programmes and projects rather than actually delving into policy itself. This brings me to the key arguments I would like to make about the NSTIP policy, in particular, and policymaking in Pakistan, in general.
As a student of public policy, particularly as it is practised in Pakistan, I have three main assertions and arguments about how policy is practised in Pakistan and why it often doesn't seem to work.
First: Policy must be minimalist. It is almost a fashion in Islamabad to issue policy documents that are at least 100 pages long. The policy documents include a thorough overview of the sector for which it is made – including paragraphs and paragraphs of meaningless text often written by a junior staffer or cut-and-pasted from somewhere – such that the actual operative part of the policy document is drowned in endless and 'meaningful' text. If our policy is not at least 100 pages long, it hardly ticks the box for being useful.
The truth is quite the opposite. Good policies should be designed to be minimalist, i.e., having the least number of words and where every word is made to count. I often think of good Constitutions as effective policy documents. The US Constitution, for example, has 4,500 or so words, and the meaning of every single word counts, and (constitutional) battles have been fought over the meanings and interpretation of these words for more than two centuries. The shorter and more impactful the policy document is, the easier it will be to implement.
Looking at the NSTIP document, it is quite clear that even though the document – for the first time – clearly lays out some key policy statements and themes, they are just too many to be implementable – about 63 in all (or 1,700 or more words). A closer look at some of these statements suggests that they offer neither a change in direction nor are even meant to be implemented, for the policy itself carries no teeth unless it is somehow enforced or funded, or both.
For example, policy statement #1 (“…key societal challenges will be defined… and address(ed) … on a regular and consistent basis”), #3 (“…increase the number of researchers per million population…”), #4 (“…annual Research and Development spending will be enhanced … to the world level…”), #6 (“…performance of researchers will assessed, recognised and rewarded…”), etc. Clearly, no substantial change in direction here. For others, such as #2 ("…availability and proper utilisation of latest equipment…"), #7 ("…access to internet will be enhanced across the country…"), #12 ("…appointment of right STI leaders in Science and Technology organisations…"), and #17 ("…National Commission for Science and Technology (NCST) shall be revived…") etc., one must ask why these are "important" policy statements in the first place? For instance, is it currently not the intention or responsibility of the Ministry of Science and Technology (MoST) to appoint the right STI leaders or make NCST function properly?
After spending time at the highest levels of the government, I learned that a very important part of government dysfunction, generally, is the turf wars between government ministries and departments
Still others such as #10 ("…ten cities be transformed into 'smart cities' by adopting the latest concepts of smart urban planning, technology integration, and service provision…"), #18 ("…Science, technology and innovation will be included in the Constitution through a constitutional amendment, and a Science, Technology and Innovation Act…"), and #28 ("…Science and technology-oriented social culture, harmonised with our religious and traditional cultural values, will be promoted…") etc., are included by either entirely a wishlist and hence unimplementable, or never meant to be implemented at all. Take the idea of doing a Constitutional amendment to reflect the importance of STI in Pakistan – S&T is already part of Schedule IV of the Constitution as a federal responsibility. The NSTIP document gives several examples where science and scientific inquiry have become a part of the constitutions of countries such as South Korea, Iran, the Philippines, the US, etc. But one must ask if these countries' care of the importance of STI came first, or did the Constitutional amendment force them to care for STI? In the case of the United States, at least, the idea of intellectual property rights – for instance – was fundamentally important to the founders of the US and they pushed to enshrine its protection in the Constitution and not the other way round. Is this an idea the people of Pakistan (or their governments) fundamentally care about? I seriously doubt it. And so, any change in the Constitution to add something that most people (aka politicians don't care about) will have the same null effect as any other (inoperative) clause the Constitution currently has.
This is not a hypothesis anymore but a living reality. The NSTIP policy has been in existence for over two and a half years now, and hardly any of these policy statements have made any difference – whatsoever – to the promotion and adoption of STI or its actual practice on the ground. Zero. Nil. Nada.
An ideal NSTIP policy – in the manner that I describe above – could have had 10-15 significant "changes" in a direction that would be implementable and extremely consequential.
There are also two corollaries of my first assertion.
The first corollary is that policies must be different from plans, programmes, or projects. They must set the broad directions in the most meaningful way (and identify the change of direction in the most dramatic way, where possible) and, identify key performance indicators (quantifiable measures of impact or changes being sought) and let the relevant ministry or departments figure out how they would implement. This will ensure that implementing entities will also have some leeway in deciding their own course of actions and hence room for performance rather than being dictated everything from the top down.
As chair of an R&D Reforms Committee, I was once asked to review the Higher Education Commission's (HEC) revised policy on Offices of Research Innovation and Commercialisation (ORIC). ORICs were set up amongst a series of initiatives taken by the HEC to improve research competitiveness in Pakistani universities, but they largely became dysfunctional, post-office-type organisations that were a burden rather than a help to research. HEC had asked a "consultant" to help them write a new ORIC policy and they produced a document spanning tens of pages that described every single thing that the ORIC must (and must not) do and how they should do it, and whom they should hire, and how much they should be paid, etc. etc.
This type of policymaking is quite non-sensical as (a) it takes away any freedom and initiative from those who are required to implement it, and (b) it is impossible to implement in the first place. What was really needed in the above instance was a short, concise, and impactful policy – no more than five pages, perhaps – that clearly described what was expected of ORICs, identified a very small number of do's and don'ts that were critical to achieving these expectations, a small number key performance indicators (KPIs), and real incentives (or disincentives) for performance (or non-performance). For example, better-performing ORICs meeting the criteria and KPIs could charge a higher percentage of overhead rate on their projects.
This brings me to the second corollary: it is very hard to successfully implement an expansive and verbose policy document. It is hard to actually implement in the sense that even before such a policy is launched, and certainly very soon after, it becomes obsolete. Unless, obviously, it has been designed precisely such that nothing concrete is aspired for, and nothing is expected to be delivered. This is also a fine art of Pakistan's bureaucratic institutions, but that's a subject for another day. For now, what is important is that the more expansive and verbose a policy is, the more likely it is that the actual practice will supersede and circumvent it. A minimalist policy with a small number of changes in direction will create a greater impact than an expansive one with lots of meaningless do's and don'ts. Our policies often fail on this count as they tend to be too prescriptive and resemble a set of detailed plans and programmes rather than broad policy statements (more on this later).
Second: The locus of policy must belong at the highest level of the government – i.e. the prime minister and his/her cabinet – with the ministries serving as implementation arms. Again, this may sound a bit counterintuitive that all policies must originate with the prime minister (and his/her cabinet) and not the minister in-charge, which is the usual practice in Pakistan. But I will argue that the usual practice in Pakistan is faulty and, in fact, a misinterpretation of the actual rules (of business).
It is true that the minister (or the ministry) in-charge is the one entrusted with implementing the policy and must have a proper say in the policymaking process, but he/she isn't the owner of the policy – the prime minister and his/her cabinet are. Our rules of business actually dictate that the cabinet is the approving body for any policy document, and so, the cabinet must also be the body that is accountable if implementation doesn't happen. Unfortunately, cabinet approvals are rubber stamps and done superficially and haphazardly, damaging the policymaking process.
After spending time at the highest levels of the government, I learned that a very important part of government dysfunction, generally, is the turf wars between government ministries and departments. Unfortunately, much more energy and effort in government is spent on undermining somebody else's performance and capturing their piece of the pie than on improving one's performance and using one's resources effectively. This turf war over influence and money reflects badly in our policy documents.
It is the job of the Prime Minister to resolve these turf issues at the level of his/her cabinet rather than kick the proverbial can down the road and have this turf war fought in the realm of stated (and implied) policy. Take the example of the National STI Policy 2022 before us. The Ministry of Science and Technology has issued policy statements and directives on matters over which it has no control at all and they, in fact, deal with the ministries of Information Technology (IT) and Telecom (such as Cybersecurity, Smart Cities, Technology Funds), HEC (PhD scholarships, resource and manpower development), Ministry of Education (STEM education, creativity of children), etc., for which these ministries are themselves responsible to the prime minister and the cabinet. These overlapping responsibilities and disparate and unconnected actions by different ministries are bound to fail to deliver impact.
Effective policymaking and implementation will require proper scrutiny and rationalisation by the federal cabinet before approval to avoid duplication of effort and the emergence of turf wars between ministries over resources.
A corollary of this is that ministers (or ministries) who have aggressive and expansive tendencies tend to dominate the country's policy (or development) agenda, make over-ambitious policy pronouncements and end up securing most of the resources, while those who are passive suffer. If the locus of policy was with the prime minister's office, the policy agenda would come to him/her, and inter-ministerial conflicts would be resolved at a higher level, with ministers becoming agents of delivery who compete on the quality of implementation.
Third: Policy must take a very critical view of the ability of the government to actively plan, execute, and oversee a series (or cascade) of actions and steps and must focus, to the extent possible, on actions that do not require direct governmental/bureaucratic involvement or repeated decision-making. This is particularly true of countries like Pakistan, where the government's capacity to execute anything effectively and transparently is severely constrained.
When a self-executing policy can be found, it is the most effective way to ensure implementation. The converse, obviously, requires the creation of additional bureaucracy, appropriating funds on a sustained basis, laying out detailed plans and assigning oversight
Through my years in government and over two decades of keenly watching Pakistan's policy landscape, I have become a lot more aware of government dysfunction and, therefore, a lot less ambitious when it comes to expecting results from policies, primarily because it is hard to back them up not just with funds but also with the ability to actively oversee, execute, and fine-tune. However, there is one notable exception here, and I call this the exception of a "self-executing policy." A self-executing policy does not require a government official or a bureaucrat to wake up in the morning and decide to make an effort to do a good job implementing a particular policy – implementation happens all by itself.
Let me give an example. In the heydays of mobile telephony in Pakistan (and elsewhere), progress was marred by the inability to find a way to bill customers. There were a myriad of schemes, including ones where those who received the call also had to pay a certain amount for every call they received. Then, in the 1990s, a literal policy change happened in Pakistan and elsewhere. It was called 'Calling Party Pays' (or CPP), and this single piece of policy created the telecom revolution in Pakistan, which has been credited with spurring technology-led economic growth and development for at least a couple of decades.
Although it may sound quite intuitive with the benefit of hindsight, the CPP was a masterstroke at the time because it did not require implementation bureaucrats day in and day out but just a simple change in the billing systems of one (or two) mobile operators. A self-executing policy with a far-reaching impact resulted, and the rest is history.
In creating policies, particularly those requiring a significant and dramatic departure from the past, one must carefully look for opportunities to find self-executing policies. It is true that often one is not possible, but when a self-executing policy can be found, it is the most effective way to ensure implementation. The converse, obviously, requires the creation of additional bureaucracy, appropriating funds on a sustained basis, laying out detailed plans and assigning oversight, and expecting timely, well-informed, and enlightened action from a set of bureaucrats which is often a non-starter and leads nowhere.
Sometimes, affecting change may require a shock or jolt to the system. Let's take, for example, policy statement #21 ("…Cognitive skills of the children will be developed, by enhancing teaching skills of science teachers, modernising syllabi and adopting novel teaching methods such as storytelling of scientific concepts, learning through playing, field trips, etc…") and #22 ("…The concept of STEM education will be implemented in select schools … to be replicated at a larger scale…"). These, attempted traditionally, will require a major effort to retrain millions of teachers, setting up curriculum committees and councils, writing new books, etc. One can pretty much write off seeing any impact of this policy statement for decades, if not never. Alternatively, a self-executing policy of abolishing textbook boards, making standardised tests, cover and test concepts rather than a specific book or physical curriculum will shake the system, invite teachers and students to struggle for a couple of years but then change course and begin learning concepts rather than cramming a set of problems at the back of an official textbook. This would probably amount to throwing the system off the cliff and letting it assemble a parachute as it falls, but it will be the most effective way to affect change. This is probably an extreme example of a self-executing policy, but hopefully, it makes the point.
A corollary of this is that to be effective, one must limit the ambition and size of the policy interventions or plans to a realistic level rather than making over-the-top claims and pronouncements at the start and finding oneself barely advanced five or ten years later. Identifying a small set of actions that are doable, likely to make the most impact and then pursuing them with laser-like focus using minimal bureaucracy is the only way to make progress where one isn't afforded the possibility of a self-executing policy.
To sum up, I'd like to once again use the example of South Korea's effective and interventionist industrial policy through much of the last half a century or more. Koreans started at a very low level, with little or no access to science or technology. Korean policymakers and planners picked the one (most) obvious sector for policy intervention, namely, textiles and garments and provided easy access to credit for the private sector to set up the textile and garments industry – a low-tech but high-employment industry. Within a decade, these incentives were phased out and Koreans moved on to encourage the light engineering industry – a natural value-addition to the low-tech textiles and garments. In the third decade, the incentives moved to heavy engineering, then to automotives, and then to electronics, and now to even more advanced industries such as mobiles and life-sciences, etc.
This is a simplification, but the general point remains valid. The Korean five-year plans were supra-ministerial initiatives that put the locus of policy where it belonged – the highest level – and everybody else fell into place implementing those plans. Policy (and plans) in Korea demonstrated a clear change in direction every decade, not the continuation of a diluted and ineffective status quo. Incentives were performance-based, temporary, and (almost) self-executing. Once the credit instruments were put into place, a bureaucrat did not have to make decisions and fine-tune every day. The availability of credit and incentives created their own momentum of implementation. And finally, taking a limited view of its ability to execute, the government did not try to do everything itself and involved the private sector in executing these plans.
Catching up with South Korea will be a tall order for the Pakistani state. However, where we could start now is to look critically at our process of policymaking and make it evidence-based, realistic and effective to bring about large-scale change in our economy and our society.