Thursday, May 19, 2016

About domestic resources and infrastructure financing in Nepal

There has been a lot of debate about the use of domestic funds to finance large-scale infrastructure projects, especially hydroelectricity, roads and airports. Let me provide a brief background and readers can see for themselves which claims made by ‘experts’ hold water and which ones are impractical and outlandish. 

First, issue about fiscal space to finance large scale projects. Nepal has a very low outstanding public debt (about 25.6% of GDP), with internal and external debt stock amounting to 9.5% and 16.1% of GDP, respectively. The outstanding public debt decreased drastically from about 52% of GDP in FY2005. This was a result of faster repayment of principal and interest (due to high revenue growth rate but eroding low expenditure capacity). Now, given the huge infrastructure deficit and low per capita income, Nepal can afford to increase it by few percentage points and use those money in productivity-enhancing infrastructure projects. Here is brief study on increasing public debt by 10 percentage points to finance post-earthquake reconstruction without jeopardizing fiscal sustainability (that is, maintaining the present level of primary balance). 

To increase this limit further needs attention on two fronts: (i) a higher growth rate (ideally above 5%), concessional nature of external borrowing and higher remittance-financed imports, which then boosts revenue; and (ii) drastic enhancement of absorption capacity. 

On the first point, no comment on growth rate as we know well what drives it in Nepal. But, the concessional financing may be drying up. The ADB is phasing out concessional lending in few years and Nepal may have to borrow at a rate between concessional rate (for low income countries) and nonconcessional rates (for lower to middle income countries). The WB is also moving in the same direction. AIIB financing may not be as concessional as those from the ADB and the WB. We will have to wait and see that one. Bilateral financing may or may not be equally generous in terms of interest and maturity, but the downside is that there are hooks on procurement and utilization of funds.

On remittance-financed imports, remittance inflows may not be as stable as before given the continued slump in oil prices and fiscal strain in the countries where Nepali workers go for employment. Remittance inflow is already decelerating, i.e. its growth rate is decreasing although the amount is rising.

On the second point, without drastic enhancement of absorption capacity, even concessional financing won’t come (disbursement happens after government spends money, submits the bills or details of spending to donors and then gets reimbursed). And it is no secret that spending absorption capacity is receding in the last couple of years. So, yes POTENTIAL financing options are available, but REALIZING it is an entirely different ball game. It is not as simple as it sounds. 

What about domestic financing (from treasury savings), domestic borrowing and using some forex reserves? Again, utilizing these is linked to macroeconomic balance and fiscal space. Given the occasional fiscal surplus (even primary balance is in surplus) the government can afford to run fiscal deficit IF the additional money is used in productivity-enhancing infrastructure projects in an accelerated manner. This is has been more and more unlikely given the receding disbursement rates of major projects and the usual legal, institutional and political hiccups overpowering project implementation. 

A lot of folks are also talking about pooling in the liquidity in the market and household savings apart from those in the banks (as seen by several IPOs being oversubscribed in recent years) to generate funds for large-scale infrastructure projects. But then this liquidity is transitory because of the adverse market conditions. Domestic financing will quickly dry up as soon as real estate and trading activities regain momentum, and BFIs see a favorable credit demand with less socio-political risk. Another aspect associated with this is the domestic borrowing by government through the selling of bills and bonds. This cannot be extended beyond 2-2.5% of GDP to maintain fiscal stability. This is already being practiced (as the government targets to raise around Rs50 billion annually). Higher domestic borrowing by government will push up lending rates (not good for private sector and households as it stifles growth and then subsequently revenue and then finally the funds available for such projects—boomerang!).

Additionally, most of the savings are for short term, but infrastructure financing is for long term. So pooling in money to explicitly finance large-scale infrastructure projects is difficult as folks don’t want to wait for years to get financial benefit. This is different from IPOs, where folks want to purchase shares with an objective to trade it at higher prices after the moratorium on selling is over after few months. Also, banks will be unwilling to invest in large amount in long-term projects out of the short-term deposits. Can you see the asset-liability mismatch there (and the troubles BFIs would be in)?

Next, what about treasury surplus? Again, it is not as easy as it sounds as the government has been using a portion of it as “carry over from last year” to finance the burgeoning (recurrent) expenditure. Not a new thing, but its size will deplete as soon as the government spends more on other projects across the country. If we pin much hope on treasury savings, then we are basically saying lets decrease capital spending in the rest of the country, save the money and then use it in few of the large infrastructure projects. The treasury savings (about Rs50 billion in the first nine months of FY2015) are there because the government is raising revenue (thanks to remittance-finance imports and consumption) that it cannot spend on in infrastructure projects. It is not an additional bonus to domestic financing that you can use for large-scale infrastructure projects! Importantly, at the core of it, it does not solve the problem: the government is not able to spend the money it appropriates for various infrastructure projects across the country. The focus should be in resolving this aspect and ensuring that major projects on an average spend over 90% of allocated money (but not 60% of those in the last quarter; preferably at least 50% by the first seven months of fiscal year).

Finally, what about the forex reserves (about $10 billion as of the ninth month of FY2016). Nepal cannot use this like that because it needs to maintain enough reserves to finance around 7 months of import of goods and services (given the peg to the Indian rupee, to fend off external shocks and the opportunity cost of holding reserves). And the amount of reserves fluctuates depending on remittance inflows and the level of import of goods and services.  

Overall, yes some room for domestic financing is there, but not as high as is being perceived by folks out there.

Tuesday, May 3, 2016

Trade blockade was more damaging to Nepalese economy than the earthquake

The Central Bureau of Statistics (CBS) just released its GDP growth projection for FY2016. The growth projection is quite close to the latest IMF projection. It offers further clarity on the impact of the trade blockade and supplies disruption (which lasted for four and a half months) and the lingering effects of the earthquake.

Overall, GDP (at basic prices) is expected to grow at 0.8% in FY2016, lower than 2.3% in FY2015 (it is a revised figure, which provisionally was estimated at 3.0% after the earthquake). Agriculture sector is expected to grow at 1.3%, industry at a negative 6.4% and services at 2.7%. 



While the earthquake affected mainly central and western regions, the trade blockade and supplies disruption severely dented economic activities across the country. In this sense, the trade blockade and supplies disruptions were more devastating than the earthquake. Furthermore, the delay in distributing cash relief and housing grant, and in initiating reconstruction projects affected growth prospect. It itself was affected by the blockade as construction materials and fuel could not be imported. The trade blockade and supplies disruptions exacerbated the already weak economic activities and squeezed economic potential. Here are earlier blog posts on this issue: one year after the earthquake; impact of trade blockade on fiscal sector, and inflation and external sector



The debilitating impact of trade blockade and supplies disruptions is visible from the negative growth rate of mining and quarrying (-6.5%); manufacturing (-10.1%); electricity, gas and water (-1.7%); construction (-4.0%); wholesale and retail trade (-1.1%), and hotels and restaurants sub-sectors (-4.9%). Furthermore, within the services sector, growth rate of transport, storage and communications sub-sector was lower than in FY2015. 

All of these sub-sectors are import-dependent either for final goods or intermediate goods or raw materials (including petroleum fuel and cooking gas). The effect of these unexpected blows will linger on in the coming months as economic activities are severely disrupted and dislocated by the double whammy of earthquake and blockade/supplies disruption. The acceleration of reconstruction work will be a major factor to stimulate economic activities and to expand potential growth.

Given the reduction in paddy, wheat and millet production, the CBS's growth projection for agriculture sector looks optimistic. I think the revised figures later in FY2017 will show even lower GDP growth rate as figures are adjusted for agriculture and some services sub-sectors. It could be one of the lowest in the last few decades.



In FY2017, a lot will depend on the weather, normalization of imports and distribution networks (including fuel and cooking gas) and the scale and pace of reconstruction activities. The El Nino conditions are already contributing to droughts and forest fires have destroyed settlements and land. The monsoon rains may not be as plentiful as expected given the conditions so far. This is going to dent agricultural sector’s recovery. The bright spot is that reconstruction work may accelerate and housing grants may be finally distributed to all the eligible households. The former will stimulate domestic production of industrial goods and services. The latter will act as a direct fiscal stimulus that will increase consumption demand.

Also, both could reinforce each other and provide the necessary jolt to the sluggish growth rate. Fast recovery of the services sector will be contingent upon the rate of remittance inflows and normalization of supplies. Overall, considering these prospects and the low base effect, GDP growth could be around 5% to 6% in FY2017. The risk to this outlook is sluggish reconstruction work (plus budget execution), delayed normalization of supplies and a slowdown in remittance inflows

By the way, revised figure for FY2015 shows that real GDP grew by 2.3% in FY2015, meaning that the earthquake had a marginally large impact on the economy than previously estimated. Earlier, FY2015 growth was estimated at 3.0%, meaning that about 1.5% of growth was chopped off from the estimated growth of 4.6% in a no-earthquake scenario. Now, its 1.6% off the estimated growth of 3.9% in a no-earthquake scenario (the difference in pre-earthquake growth estimate is due to the updated actual figures for FY2014).

Structural change in China and its impact on other economies

Asian economies face headwinds from three weak factors: weak global recovery, weak global trade, and weak (rather moderating) Chinese growth. To deflect such headwinds, the economies need to focus on productivity-enhancing structural reforms, which are needed to rebalance demand-supply, reduce vulnerabilities, enhance efficiency, and to expand the frontier of potential growth. Meanwhile, monetary policies need to be harmonized not only in Asia, but among advanced economies as well.

What could be the potential spillovers from China’s economic transition toward a more sustainable growth? Note that China moving on this path (means living with slower growth than in the past years) would mean slower demand for raw materials and intermediate goods, which in turn means trouble for economies whose activities survive on exporting such goods to China (they are also facing slow global recovery [plus low commodity prices], which in turn is slowing down demand for final goods either made or assembled in China). Also China is beginning to produce more intermediate goods rather than importing them for use in final goods as before. The rebalancing of the sources of growth in the Chinese economy has consequences beyond its borders. 

China accounts for about one-half of regional growth and is an important factor in the regional supply chains.

According to the latest regional economic outlook, the IMF argues that:
  • Asia has become more sensitive to the Chinese economy (which is moving away from high public investment and exports to higher domestic consumption and services). The rebalancing of Chinese economy will have adverse short-term effects in the rest of Asia. This may be less painful for economies that respond to Chinese consumption demand. 
  • Estimates show that a 1 percentage point slowdown in Chinese growth translates into a 0.15–0.30 percentage point decline in growth for other Asian countries in the short term. Taiwan, South Korea and Malaysia are likely to be affected the most.
  • Commodity exporters such as Australia, Indonesia, Malaysia and New Zealand will suffer from slowdown in Chinese economy (as it imports a lot from them) and lower global commodity prices.
  • Disruption to the financial sector in China also poses risks to the global economy. Rapid credit growth, high levels of NPLs, large shadow banking, and growing corporate and local government debt may be disruptive and could affect cross-border financial linkages (forex markets, trade insurance, equity markets, etc)
Apart from the China factor, some economies also face economic and fiscal strain from natural disasters such El Nino and cyclone. El Nino conditions have worsened drought and lowered agricultural output in many economies, including Nepal and India.
 
The IMF recommends:

Although the global economic panorama remains turbulent, policymakers in Asia will need to continue to build on the region’s strengths. Harnessing Asia’s potential will call for strong implementation of a wide-raging policy agenda, including enhanced communication of policy frameworks and goals. Structural reforms, aided by fiscal policy, should support economic transitions and bolster potential growth. Monetary policy should remain focused on supporting demand and addressing near-term risks, including from large exchange rate depreciations and deflationary shocks. Policies to manage risks associated with high leverage and financial volatility will play an important role, including exchange rate flexibility, targeted macroprudential policies, and in some cases, capital flow measures. Finally, policy recalibration should not lead to a buildup in vulnerabilities.


Thursday, April 14, 2016

Nepal was the third largest remittance recipient in 2014

As a share of its GDP, Nepal was the third largest recipient of remittances in 2014 (see the latest Migration and Development brief by WB). Remittance inflows to Nepal amounted 29.2% of GDP in 2014. Tajikistan and Kyrgyz Republic received remittances amounting to 36.6% of GDP and 30.3% of GDP, respectively. In 2015, remittance inflows to Tajikistan and Kyrgyz Republic was $2.6 billion and $1.7 billion, respectively. In Nepal, it is estimated to be about $7 billion. 

The remittance inflows soared last year immediately after the earthquake and has been one of the most important income sources (sans faster government aid) of the earthquake-affected households. Increased inflows to finance daily household needs has been one of the most vital coping strategies of the affected households.



In 2015, Nepal received remittances from around 35 countries. The highest bilateral remittance inflows was from Qatar ($2.02 billion), followed by Saudi Arabia ($1.8 billion), India ($1 billion), UAE ($803 million) and the United States ($332 million). Qatar and Saudi Arabia absorbed about 124,368 and 98,246 migrants respectively in 2015. Malaysia was the most popular destination with 202,828 migrants in 2015, but the remittance inflows from Malaysia are lower than from other destination ($185 million). This probably might be due to the lower wages and less working hours plus the informal inflows. Or/and, the WB's estimation might have simply missed the realistic figures. For instance, remittances from Japan to Nepal are shown to be zero, but the money transfer agencies are doing brisk business (eg. Kyodai/IME Japan). Most of the countries waived remittance costs immediately after the earthquake, leading to a surge in inflows. 

Interestingly, India received $2.7 billion as remittances from Nepal in 2015. These bilateral remittance estimates are computed using data on migrant stocks, host country incomes and origin country incomes. Here is a cautionary note from the authors: "These are analytical estimates based on logical assumptions and derived from a global estimation of bilateral remittance flows worldwide. They are not actual officially reported data. The caveats attached to these estimates are: (a) the data on migrants in various destination countries are incomplete; (b) the incomes of migrants abroad and the costs of living are both proxied by per capita incomes in PPP terms, which is only a rough proxy; and (c) there is no way to capture remittances flowing through informal, unrecorded channels."


Here is a chart showing the migration and remittances trend after natural disasters.


Wednesday, April 13, 2016

Reality and requirements for Nepal-China accords to work

Nepal needs serious homework to benefit from accords with China, including the important one on transit

The agreements signed between Nepal and China during Prime Minister KP Oli’s state visit has in theory ended the sole dependence on India for utilizing the transit right of a landlocked country. Furthermore, theoretically it has also increased the odds of procuring petroleum fuel and other essential supplies either from or through China. It marks a remarkable symbolic departure from the exclusive foreign, trade and transit relations Nepal has had with India for decades. The impetus for this was triggered by the supplies disruption between Nepal and India, and the latter’s cold reception and subsequent reaction to the promulgation of a new constitution last year.

Practically, it bears little significance unless Nepal upgrades existing connectivity as well as constructs new commercial custom points with China, reduces cost of doing business, establishes trust among traders on both sides, and boosts productive capacity by taking decisive action on policy and implementation fronts. 

New agreements

Nepal and China released a 15-point joint statement, which was missing during PM Oli’s state visit to India, detailing the new understanding and agreements on a range of issues. The most consequential one is the Agreement on Transit Transport, whose operational details (protocol of the agreement) are yet to be worked out. The Chinese government will also “seriously consider” to provide enhanced market access to tradable goods and start work on joint feasibility study of Nepal-China Free Trade Agreement. It also includes promises to conclude a commercial deal on the supply of petroleum products. 

Frustrated by the acute shortage of essential goods and supplies for about five months, analysts and the general public were quick to extol the agreements and are hoping for uninterrupted supplies and unhindered trade. However, operationalizing such agreements is not quick and easy. It requires serous homework, especially on our side, on policy, infrastructure, financial and procedural fronts. The usual tardiness of bureaucracy and meddling politicians will potentially slowdown implementation.

Trade and investment

The open border, state of infrastructure, business and family relations, language, free movement of labor, and the currency peg form the bedrock of Nepal’s trade relations with India. These are missing in the case with China and hence implementation will be even slower and difficult.

India accounted for 65.5 percent of total export and 63.5 percent of total import in FY2015. The figures for China are 2.6 percent and 12.9 percent, respectively. Accordingly, balance of trade with India was 62.2 percent of total trade deficit and 14.2 percent with China. The currency peg, which has remained unchanged since 1993 and has generally fared well for Nepalese economy given the fragile economic and infrastructure fundamentals, has supported the large and growing trade with India. Nepal’s top exports to India are light manufacture goods such as textiles, polyster yarns, zinc sheet, jute goods and some agro-processed items like juice. 

The Nepal India trade treaty allows for duty free access of manufacturing goods to India, a major incentive for Nepalese exporters to continue production despite the relatively high cost. Similar tariff preferences are not available with China and the duty free access it allows to its market is applicable to all least developed countries (LDCs) as per its commitment during the global trade negotiations. Nepalese exporters face a relatively high transaction cost and tough competition in the Chinese market. Tanned skin, handicraft, woolen carpet and noodles are the top export items to China. Interestingly, these light manufacture goods are produced by importing intermediate goods from India itself. This dynamics is not going to change anytime soon given the industrial base and its sophistication.

Regarding imports, the most important one is petroleum product, which accounts for about 20 percent of total import and is higher than the total value of merchandise exports. India has been the sole supplier of petroleum fuel, which is the largest import item followed by vehicles and spare parts, rice and paddy, and other machinery parts used in Nepal’s industrial sector. Meanwhile, Nepal’s top imports from China are telecommunication equipment, electrical goods and chemical fertilizers. 

Looking at the existing composition of export and import, it is not hard to notice that the basic items required by households and business community are actually imported from India and given the state of infrastructure, exchange rate regime, financial and business connectivity, it is not going to change overnight. For instance, due to the long distance and rugged terrain as well as Chinese taxes, importing fuel from China is about two times expensive than importing from India. 

If we look at investment, Indian investment is concentrated in manufacturing and energy sectors, while Chinese investment is focused on energy and services sectors. Overall, Indian investment is higher than Chinese investment. Meanwhile, Indian airlines bring in the largest number of visitors to Nepal, but the share of Indian and Chinese tourists is 17.1 percent and 15.7 percent, respectively with the latter growing at the fastest rate. Indian and Chinese foreign aid commitment is about 9.2 percent and 3.5 percent, respectively, of total aid commitment.

The other most important aspect is the open labor market in India, which absorbs a majority of the seasonal migrants from poor households from upper part of far-west and mid-west and the Terai belt. The remittances from India is an important source of the household’s expenditure. This access is missing in the case of China. Proximity-wise and cost-wise, the Indian market will continue to be more attractive than the Chinese market.

Ensuring viability

How can Nepal benefit from the renewed rapprochement with China on political-economic front, but also gradually lessen the over dependence on India? Given the state of infrastructure, trade and investment pattern, and business relations, there is little to gain in the short term. Like it or not, Nepal’s dependence on India for trade, investment and third country access will not decrease any time soon. In the medium to long run, a major effort to match the infrastructure, financial and business connectivity with China could be a game changer. 

However, this is easier said than done given the bureaucratic and political tardiness in implementing major infrastructure projects. Nepal has a lot of internal homework to do in terms of large public investment in infrastructure for enhanced connectivity, energy generation, capacity development, and country-specific export target based on Nepal’s comparative advantage.

Else, the agreements with China will bear no meaning beyond symbolism and we will continue to depend on India for pretty much everything used by households and businesses. The Chinese market is open, but is not easy to access and penetrate. Nepal is in a long haul to operationalizing the agreements with China and benefit from it.


It was published in The Kathmandu Post, 11 April 2016

Friday, April 8, 2016

One year since the Great Gorkha earthquake

It is nearly a year since the catastrophic 7.8 magnitude earthquake struck Barpak, Gorkha. The massive earthquake on 25 April 2015 at 11:56 AM and subsequent aftershocks (two powerful 6.7 magnitude on 26 April and 7.3 magnitude on 12 May) caused widespread damage to lives and properties. About 9,000 were dead and 22,000 injured. Around 602,257 and 285,099 private houses were fully and partially damaged, respectively, forcing thousands of people to seek temporary shelter under tents and tarpaulin sheets. 

It was a natural disaster most were expecting, but the government and households were awfully unprepared— the government either had lax household and settlement policies, or/and did not enforce housing codes strictly; the households (mostly in urban areas) compromised safety standards to save money. Lessons were learnt or so the politicians, bureaucrats and the public said during the first few months of the disaster. Then the simple lessons inculcated only after a terrible tragedy evaporated in thin air as politicians paid lip service only, bureaucracy got mired in the usual maze infested with rent-seeking mentality, and the affected households did what they need to do to get by normal lives (by starting rebuilding houses on their own). It’s a sad story about reconstruction that did not happen even after a year into the disaster. Availability of funds was not a problem given the generous aid pledges during reconstruction conference.


First, a quick snapshot of the events thus far: A post disaster needs assessment (PDNA) was completed within two month. It estimated the cumulative damage and loss to be about $7.1 billion (33.3% of FY2015 GDP). The cumulative need for recovery (Building Back Better concept) was estimated to be $6.7 billion, of which almost half was needed for housing and settlement. The earthquake either destroyed or damaged cultural heritage, schools, health posts, houses, agricultural farms, irrigation system, financial sector infrastructure, communications, roads, hydroelectricity plants, livelihoods, and tourism infrastructure, among others. 

After the completion of PDNA, the government organized International Conference on Nepal’s Reconstruction (ICNR) on 25 June 2015 to raise funds for reconstruction. Bilateral and multilateral development partners made generous pledges to help the affected households get back on feet again. About $4 billion, which was larger than the estimated public sector needs, was pledged with a belief that the government would speed up the establishment of reconstruction authority and make it operational without delay. The initial expectation was that some form of relief would be distributed before the winter. This was to be followed by cash grant of about $2000 to rebuild destroyed houses so that some of the damaged houses could be rebuilt before the arrival of monsoon (June to September 2016). Then the expectation was that the government would plan resettlement where required and build back better the damaged infrastructure, which would help reinvigorate the economy, create employment opportunities and prepare the household better when the next disaster strikes (Nepal is in a seismically active zone). Alas, these remained unrealized expectations. More on this in a minute.

  • Total damage estimated at $7.1 billion. Total recovery needs estimated at $6.7 billion. 
  • Total reconstruction aid pledged by bilateral and multilateral donors was around $4 billion.
  • About 9,000 dead and 22,000 injured.
  • About 602,257 and 285,099 private houses were fully and partially damaged, respectively.
  • About 2,673 and 3,757 public buildings were fully and partially damaged, respectively.
  • Earthquake struck in the tenth month of FY2015. The earthquake chopped 1.5 percentage points off an estimated 4.6% growth in FY2015 in a no-earthquake scenario. GDP growth and per capita GDP estimated at 3.0% and $762 in FY2015. Services sector was hit the hardest.
  • An additional 700,000-982,000 people were pushed below the poverty line. This translates into an additional 2.5%-3.5% of the estimated population in 2015 pushed into poverty compared to the no-earthquake baseline scenario of about 21%. About 50%-70% are from rural central hills and mountains, where the vulnerability prior to the earthquake was already high.
  • An ordinance was promulgated to establish National Reconstruction Authority (NRA) and to start reconstruction activities. 
  • The FY2016 budget and monetary policy were focused on rehabilitation and reconstruction. About $910 million (3.8% of GDP) was earmarked for reconstruction related work ($740 million for the NRA and $170 million to be spent through sector ministries during the interim period, i.e. till the NRA was operational).
  • The FY2016 budget included NRs200,000 (about $2000) for each household that has lost its house due to the earthquake. To address the shortage of labor for reconstruction, skill training is planned for 50,000 people in the areas of masonry, plumbing, and electrical works.

 
Now, nearly a year since the earthquake, the progress is rather disappointing. Few points:
  • The political wrangle between CPN-UML and NC to appoint the CEO of NRA delayed the passage of the NRA bill. During the NC’s time in the government (coalition with CPN-UML) it had appointed a CEO. However, when the CPN-UML led the coalition government, it demanded that a new CEO be appointed. The disagreement over this persisted for about two months. Finally, after eight months the NRA bill was passed (creating too many committees and layers within it) and a new CEO was appointed. 
  • The NPC handled the work of the NRA while the political infighting was ongoing. It approved many programs, which the NRA has asked ministries to get them approved again. 
  • The NRA is struggling to get things done on time. Even the Prime Minister, who himself chairs the executive committee within NRA, has come down heavily on the NRA for its snail pace work.
  • The NRA is not getting enough human resources and coordination from the bureaucracy to start the preliminary program, especially identification of victims and distribution of cash grant for house reconstruction. Senior bureaucrats have shown strong disinclination to transfer to NRA
  • At the local level, the political parties are trying to usurp the cash distribution process by demanding representation in local committees, under whose recommendation victims are to be identified. This is for the sake of “ownership”. There are also cases of inflated number of victims (dummy victims).
  • The CEO is not taking decisive action despite having unparalleled powers as per the NRA Act. Seeking the presence of political leaders for cash distribution in far off villages is not logical. In one case, cash distribution was delayed because some political leaders were not available.
  • The earthquake affected households have been expecting the promised cash grant for housing for almost a year now. About five months of supplies disruption has not only escalated prices of housing materials, but also created severe shortage of essential supplies in the market. The raw materials used for constructing the houses are in short supply and those that are available are sold at inflated prices. In this respect, the earthquake affected households are even requesting the government to build the houses instead of grant. They are willing to contribute labor for such reconstruction. See this and this episode of BBC Sajha Sawal. 
  • The trade blockade/supplies disruption negatively affected reconstruction planning and works. Its impact will linger on till the supplies and prices normalize to previous levels.
  • There is inadequate coordination among I/NGOs for relief distribution. Affected households are complaining of similar goods (blankets and tarpaulin sheets, etc) being distributed by many I/NGOs. And distribution is centered in some places only. The NRA has to coordinate this in the strict sense.
  • Reconstruction hasn’t started. Even a consolidated planning for reconstruction is not done yet. 
  • There is a possibility that the NRA won’t be able to utilize all the funds pledged during the ICNR last year. The more delay in utilizing the funds, the higher the changes of it being restructured for use in other purposes.
The lingering impact of damages caused by the earthquake and supplies disruption will suppress potential growth and employment opportunities. 
  • GDP growth in FY2016 was initially projected on the higher side because of the expectation of speedy cash distribution and initiation of reconstruction projects. This impact of the trade blockade for about five months and the delayed reconstruction activities will depress growth prospects in FY2016 and beyond. The agricultural output forecast is also disappointing. Some households in the earthquake affected districts have not cultivated their fields due to the uncertainty over housing.
  • Inflation will also continue to be at elevated levels, primarily due to the lingering impact of the supplies disruption.
  • There is a slowdown in the rate of overseas migration because of the cooling of demand for workers in GCC (this in turn is due to the continued low oil prices). Malaysia is implementing measures to discourage foreign workers and is also levying extra taxes on income. This means a slowdown in migration rate and ultimately deceleration of remittance inflows. It will increase the number of unemployed youths in the economy and also put pressure on external sector stability. Current account surplus will decline and if imports rebounds to normal levels, then it may as well be negative.
  • Rehabilitation and reconstruction should primarily aim at increasing productivity-enhancing public capital investment. This is a key to ensuring structural transformation whereby high value-added and high-productivity sectors are more dominant than low value-added and low-productivity sectors in the medium term. Promoting agribusiness, industrial capacity, innovation and high-productivity services need to be at the center of such a reconstruction and structural transformation strategy. In addition to higher investment, this will require reforms on institutional, legal, regulatory, and capacity enhancement fronts.
Overall, the earthquake, the severe supplies disruption and the delay in starting reconstruction projects do not bode well for the economy. It a combination of misplaced politics, ineffective bureaucracy and a lost NRA (which will try to accelerate processes in the next two weeks to show that things are moving ahead despite it being slow-- a common escape argument popular in the bureaucracy!)

Thursday, March 10, 2016

The use and abuse of p-values and statistical significance

Here is the American Statistical Association's (ASA) official policy statement on p-values and statistical significance.


Introduction

Increased quantification of scientific research and a proliferation of large, complex datasets in recent years have expanded the scope of applications of statistical methods. This has created new avenues for scientific progress, but it also brings concerns about conclusions drawn from research data. The validity of scientific conclusions, including their reproducibility, depends on more than the statistical methods themselves. Appropriately chosen techniques, properly conducted analyses and correct interpretation of statistical results also play a key role in ensuring that conclusions are sound and that uncertainty surrounding them is represented properly.

Underpinning many published scientific conclusions is the concept of “statistical significance,” typically assessed with an index called the p-value. While the p-value can be a useful statistical measure, it is commonly misused and misinterpreted. This has led to some scientific journals discouraging the use of p-values, and some scientists and statisticians recommending their abandonment, with some arguments essentially unchanged since p-values were first introduced.

In this context, the American Statistical Association (ASA) believes that the scientific community could benefit from a formal statement clarifying several widely agreed upon principles underlying the proper use and interpretation of the p-value. The issues touched on here affect not only research, but research funding, journal practices, career advancement, scientific education, public policy, journalism, and law. This statement does not seek to resolve all the issues relating to sound statistical practice, nor to settle foundational controversies. Rather, the statement articulates in non-technical terms a few select principles that could improve the conduct or interpretation of quantitative science, according to widespread consensus in the statistical community.

What is a p-value?

Informally a p-value is the probability under a specified statistical model that a statistical summary of the data (for example, the sample mean difference between two compared groups) would be equal to or more extreme than its observed value.

Six principles of p-value:

1. P-values can indicate how incompatible the data are with a specified statistical model.

A p-value provides one approach to summarizing the incompatibility between a particular set of data and a proposed model for the data. The most common context is a model, constructed under a set of assumptions, together with a so-called “null hypothesis.” Often the null hypothesis postulates the absence of an effect, such as no difference between two groups, or the absence of a relationship between a factor and an outcome. The smaller the p-value, the greater the statistical incompatibility of the data with the null hypothesis, if the underlying assumptions used to calculate the p-value hold. This incompatibility can be interpreted as casting doubt on or providing evidence against the null hypothesis or the underlying assumptions

2. P-values do not measure the probability that the studied hypothesis is true, or the probability that the data were produced by random chance alone.

Researchers often wish to turn a p-value into a statement about the truth of a null hypothesis, or about the probability that random chance produced the observed data. The p-value is neither. It is a statement about data in relation to a specified hypothetical explanation, and is not a statement about the explanation itself.

3. Scientific conclusions and business or policy decisions should not be based only on whether a p-value passes a specific threshold.

Practices that reduce data analysis or scientific inference to mechanical “bright-line” rules (such as “p < 0.05”) for justifying scientific claims or conclusions can lead to erroneous beliefs and poor decision-making. A conclusion does not immediately become “true” on one side of the divide and “false” on the other. Researchers should bring many contextual factors into play to derive scientific inferences, including the design of a study, the quality of the measurements, the external evidence for the phenomenon under study, and the validity of assumptions that underlie the data analysis. Pragmatic considerations often require binary, “yes-no” decisions, but this does not mean that p-values alone can ensure that a decision is correct or incorrect. The widespread use of “statistical significance” (generally interpreted as “p ≤ 0.05”) as a license for making a claim of a scientific finding (or implied truth) leads to considerable distortion of the scientific process.

4. Proper inference requires full reporting and transparency.

P-values and related analyses should not be reported selectively. Conducting multiple analyses of the data and reporting only those with certain p-values (typically those passing a significance threshold) renders the reported p-values essentially uninterpretable. Cherry-picking promising findings, also known by such terms as data dredging, significance chasing, significance questing, selective inference and “p-hacking,” leads to a spurious excess of statistically significant results in the published literature and should be vigorously avoided. One need not formally carry out multiple statistical tests for this problem to arise: Whenever a researcher chooses what to present based on statistical results, valid interpretation of those results is severely compromised if the reader is not informed of the choice and its basis. Researchers should disclose the number of hypotheses explored during the study, all data collection decisions, all statistical analyses conducted and all p-values computed. Valid scientific conclusions based on p-values and related statistics cannot be drawn without at least knowing how many and which analyses were conducted, and how those analyses (including p-values) were selected for reporting.

5. A p-value, or statistical significance, does not measure the size of an effect or the importance of a result.

Statistical significance is not equivalent to scientific, human, or economic significance. Smaller p-values do not necessarily imply the presence of larger or more important effects, and larger pvalues do not imply a lack of importance or even lack of effect. Any effect, no matter how tiny, can produce a small p-value if the sample size or measurement precision is high enough, and large effects may produce unimpressive p-values if the sample size is small or measurements are imprecise. Similarly, identical estimated effects will have different p-values if the precision of the estimates differs.

6. By itself, a p-value does not provide a good measure of evidence regarding a model or hypothesis.

Researchers should recognize that a p-value without context or other evidence provides limited information. For example, a p-value near 0.05 taken by itself offers only weak evidence against the null hypothesis. Likewise, a relatively large p-value does not imply evidence in favor of the null hypothesis; many other hypotheses may be equally or more consistent with the observed data. For these reasons, data analysis should not end with the calculation of a p-value when other approaches are appropriate and feasible.

Other approaches

In view of the prevalent misuses of and misconceptions concerning p-values, some statisticians prefer to supplement or even replace p-values with other approaches. These include methods that emphasize estimation over testing, such as confidence, credibility, or prediction intervals; Bayesian methods; alternative measures of evidence, such as likelihood ratios or Bayes Factors; and other approaches such as decision-theoretic modeling and false discovery rates. All these measures and approaches rely on further assumptions, but they may more directly address the size of an effect (and its associated uncertainty) or whether the hypothesis is correct.

Conclusion

Good statistical practice, as an essential component of good scientific practice, emphasizes principles of good study design and conduct, a variety of numerical and graphical summaries of data, understanding of the phenomenon under study, interpretation of results in context, complete reporting and proper logical and quantitative understanding of what data summaries mean. No single index should substitute for scientific reasoning.