Hydropower projects with PPA in dollars in Nepal

Given the long load-shedding hours, the inability of domestic BFIs to fund large hydropower projects, and the government’s limited (usable) fiscal space considering the scale and scope of investment needed, there is no doubt that foreign investment is essential to generate and supply enough hydroelectricity that can eventually revitalize the economy (through supply of adequate power domestically) and also export surplus to India (would be a good source of revenue).

However, the main sticking point for any deal is power purchasing agreement (PPA) in dollars (or foreign currency). Investors want PPA in foreign currency to avoid foreign exchange risks as they have to raise money outside and then repay it in foreign currency after generating revenue from investment projects in Nepal. It is also related to financial and institutional soundness of NEA, the dysfunctional and politicized institution that is at the center of all deals. No investor would ideally want to make a deal with a bankrupt and institutionally weak entity. Hence, the other sticking point related to this one is sovereign guarantee. Again, the investors see it as an important assurance of returns to their multi-year investments, but the government is not ready to offer such guarantee (technically, the government cannot provide sovereign guarantees for SOE’s borrowings/investments expect for the purchase of aircraft— however, exceptions have been made such as the guarantee of repayment of NOC loans to EPF and CIT). The government argues that since NEA has not defaulted so far, the government is implicitly guaranteeing its agreements anyway. However, from investor's perspective, they want credible assurances embedded in legal framework.

NEA has a bad experience with PPA signed in dollars (foreign currency). The two such projects in operation with PPA in dollars are Khimti Hydropower (60 MW) and Upper Bhotekoshi (45 MW). NEA spends around 40% of its revenue in payments to these two hydropower projects. NEA feels the heat particularly when Nepalese rupee depreciates with respect to the US dollar. Here is a nice article published on Republica (by Rudra Pangeni). Below is a list of projects, sourced from the Republica article, with which NEA has signed PPA in dollars.

NEA recently came up with a hybrid scheme: PPA in dollars for a certain percent of total hydroelectricity generated, and the rest in local currency. For the 82 MW Lower Solu project, NEA will pay only 55% in US dollar and the rest 45% in local currency. Currently, the leadership at the Ministry of Energy is against signing PPA in dollars (even some parliamentarians are opposed to it) without first assessing the liabilities to be incurred by NEA over the years. 

NEA has fixed the PPA rate for run-of-the-river projects under 25 MW at Rs 8.40 per unit during the dry season, and Rs 4.80 per unit during the wet season. The PPA rate for projects bigger than 25 MW is set after negotiations between NEA and developers.

Now, Nepal needs foreign investment (plus technology and know-how) to undertake sizable hydroelectricity projects. Investors ideally want PPA in dollars and sovereign guarantee. But, some at the leadership position are neither willing to do both nor are credibly committed to the needed unpalatable reforms at NEA (and the entire energy sector). It is up to the government to find an equilibrium that can accommodate these diverging views, and move ahead with an urgency to sustainably exploit the natural resources to power up houses and manufacturing plants.

Four binding constraints to growth in Nepal

The government and MCC have jointly published the constraint analysis report, which identifies four main binding constraints to economic growth in Nepal:

1. Policy implementation uncertainty
2. Inadequate supply of electricity
3. High cost of transport
4. Challenging industrial relations and rigid labor regulations

The analysis draws on the ‘growth diagnostics’ methodology developed by Hausmann, Rodrik and Valesco in Harvard, and builds on an earlier similar study jointly done by ADB/DFID/ILO, which also found similar constraints to economic activities in Nepal.  I had also followed the same methodology in 2009 and came up with similar constraint (mainly inadequate supply of infrastructure, including transport). 

The constraint analysis points out that protracted political transition and instability results in policy implementation uncertainty, rigid labor regulations and challenging industrial relations, and reduced government effectiveness and capital expenditures (the last one in turn leads to inadequate supply of electricity and high cost of transport). 

Below are the major highlights of the report:

---

Policy implementation uncertainty: Frequent changes in government leadership have resulted in policy implementation that has been unpredictable for firms in Nepal. While much of Nepal’s bureaucratic structure and policy documents have remained the same, changes in leadership of a ministry often leads to significant shifts in the implementation of government policy. This lack of continuity and predictability of policy implementation is consistently cited by firms as a major constraint to making investments in Nepal.

Inadequate supply of electricity: Nepal suffers from the worst electricity shortages in South Asia. Only half of the demand for electricity can be met by the nation’s grid. This results in load shedding of up to 18 hours a day during the dry winter months, when hydropower generation is low. The low availability of electricity creates significant costs for businesses which have to run generators on expensive imported fuel.

High transport costs: Nepal ranks 147th out of 155 countries in the Logistics Performance Index (World Bank LPI). While Nepal’s rugged terrain and landlocked geography contribute to this poor performance, the high of cost transportation in Nepal is also driven by poor quality and quantity of roads, a lack of competitiveness in the trucking sector, and by costly customs procedures. The result is that transporting goods within Nepal and reaching international markets is expensive and unreliable

Challenging industrial relations and rigid labor regulations: Nepal’s labor code is complex. Implementation of the code and mediation by the government between labor and business is both challenging and inadequate. The Federation of Nepalese Chambers of Commerce and Industry (FNCCI) Employers’ Council summary report identifies three primary reasons why the labor code needs revision: poor implementation, protracted court rulings, and long firing. These difficulties appear to alter the hiring and firing practices of firms in costly ways that include firm size remaining small to avoid the difficulties of labor negotiations. However, evidence from focus group discussions in Nepal suggest that these issues are improving and thus the team has categorized this constraint as less severe

---

District-wise revenue, expenditure and fiscal space

Kathmandu district contributes about 31% of total revenue but spends about 45% of total expenditure. FCGO notes that some of the development expenditures in other districts have been loaded in Kathmandu district because of difficulties in recording it in other districts—this may inflate the total expenditure figure in Kathmandu district to some extent.

Anyway, district-wise fiscal gap shows that only seven districts had fiscal surplus in FY2012, namely Parsa (Rs 59.2 billion), Lalitpur (Rs 32.2 billion), Rupandehi (Rs 11.5 billion), Morang (Rs 9 billion), Bara (Rs 3.2 billion), Chitawan (Rs 1.7 billion), and Sindhupalchok (Rs 1.5 billion). Parsa contributed about 26% of total revenue, but spent only 1% of total expenditure.

In terms of per capita revenue, Parasa topped the list in FY2012 with Rs103,728, followed by Lalitpur (Rs 86,668), Kathamandu (Rs 41,572), Rupandehi (Rs 18,567), and Morang (Rs 15,123). The lowest per capita revenue was in Bajhang (Rs 164), Achham (Rs 159), Baitadi (Rs 158), Dailekh (Rs 136), and Jajarkot (Rs 124).

In terms of per capita expenditure,  Manang topped the list in FY2012 with Rs 92,205, followed by Kathmandu (Rs 88,940), Mustang (Rs 60,364), Dolpa (Rs 27,889), and Humla (Rs 20,163). The lowest per capita expenditure was in Nawalparasi (Rs 4,498), Siraha (Rs 4,273), Bara (Rs 4,041), Rautahat (Rs 3,472), and Sarlahi (Rs 3,463).

Looking at per capita district budget deficit, the same seven districts having overall surplus also had per capita budget surplus. Per capita budget surplus in Parsa was Rs 98,488, followed by Lalitpur (Rs 68,818), Rupandehi (Rs 13,066), Morang (Rs 9,297), Sindhupalchok (Rs 5,122), Bara (Rs 4,645), and Chitawan (Rs 2,937). Per capita budget deficit was the highest in Manang (Rs 90,683), Mustang (Rs 58,672), Kathmandu (Rs 47,367), Dolpa (Rs 27,302), and Humla (Rs 19,803).

What about the relationship between district-wise per capita fiscal position and poverty rate? The basic idea is that technically districts with more fiscal space are able to spend more on the local development of physical and social infrastructures, thereby stimulating local economic activities and reducing poverty. In general, there seems to be an overall negative relationship, i.e. districts with relatively more fiscal space (in absolute terms, only seven districts had budget surplus) tend to have lower poverty rate. Now, the obvious outlier is Parsa. However, since most of the districts are having budget deficits for a long time, its impact on poverty rate might be lower (the other obvious candidates are remittances, wages, and provision of physical and social infrastructures, including social protection). The logic that relatively better fiscal space will mean lower poverty may not be as straight forward because of the severe constraint to higher absorption capacity and the lack of conducive investment climate, including energy and connectivity.

Though the proportion of expenditure in the remote districts in low, in terms of per capita expenditure (and also per capita district budget deficit), it is pretty high, indicating the high cost of delivering physical and social infrastructures in these regions. This necessitates connectivity, especially road, to reduce such costs and boost economic activities by enhancing competitiveness of niche products. Note that the districts do not have freedom to spend the revenue they mobilize. The districts recommend priority projects, which the NPC evaluates and gives a go-ahead for inclusion in the budget. The district-wise fiscal space argument is just for observation purpose.

Here is an earlier blog post on district-wise GDP and poverty.

GDP by district and region in Nepal

What is the size of regional economies in Nepal? Better what is the size of each district in terms of economic activities and what is its share in total GDP? Such disaggregated data is pretty hard to get as the CBS does not publish them in its annual (and now quarterly) national accounts estimate. Well, this blog post will shed some light on this front by distilling the share of each district and region in the national economy.  The Nepal HDR 2014 estimates the district and regional level GDP data using a variety of proxy indicators and survey datasets. Again, these are only estimates, which is the closest you can get to right now in pursuit of regional and district level disaggregated data. Earlier, a NRB study estimated that Kathmandu Valley contributed between 23.4% and 31% to GDP. This new estimate puts the figure at 20%. Data on poverty by district is covered here. District-wise household size, population growth rate, migrant population and remittance inflows data here.

Being the industry and services sectors hub of the nation, Kathmandu district’s share of total GDP is the highest—at 15.8%. It is followed by Morang (3.9%), Bara (3.3%), Jhapa (3.2%), Rupandehi (3.2%), Lalitpur (2.9%), Sunsari (2.7%), Kaski (2.5%), Nawalparasi (2.4%) and Parsa (2.4%).  The five districts with the lowest share are Mugu, Humla, Dolpa, Mustang and Manang (all 0.1% except 0.2% for Mugu).

Decomposing GDP by geographic region, Hills contribute to 48.8% of GDP, followed by 45.6% by Tarai and the rest 5.6% by Mountains. By development region, central development region contributes the most to GDP-- 45% (not surprising since the major industrial and services activities happen in this region), followed by eastern development region (20.6%), western development region (17.6%), mid-western development region (10.4%) and far-western development region (6.3%).

In terms of share of total agriculture sector, the Tarai districts top the list as this region is also the breadbasket of Nepal. Jhapa district’s share of total agriculture sector is 4%, followed by Morang 3.6%, Kailali 3.2%, Bara 3.1% and Rupandehi 3%. The lowest contributors are the districts in the mountain region.

Tarai region contributes 51.2% of total agriculture sector, followed by Hills 40.7% and Mountains 8.1%. In terms of development region, central development region is the top contributor (29.8%) followed by eastern development region 26.7% and western development region 20.2%.

Regarding industry sector, Kathmandu district contributes the highest (13.9%), followed by Bara (9%), Morang (5.3%), Nawalparasi (4.2%) and Dhanusha (4.1%). It is concentrated in Tarai region. The lowest contributors are from the mountains region. The Tarai region contributes 52.4%, Hills 43.3% and Mountains 4.4%. The central development region’s share is 53.7%.

In services sector, Kathmandu contributes 27.6%, followed by Lalitpur 4.1%, Morang 3.7%, Chitawan 3.4% and Kaski 3.3%. Tarai region’s share of total service sector is about 57.1%, followed by Tarai 38.9%, and Mountains 4.1%. The contribution of central development region is 54.7%.

Let us look the district-wise share of major sub-sectors in total of those sub-sectors (agriculture sector covered above) that drive GDP: manufacturing; construction; retail and wholesale trade; and real estate, renting and business activities.

Bara and Kathmandu districts contribute 18% and 10.7%, respectively, of total manufacturing activities, followed by Nawalparasi and Morang (7.6% each), Dhanusha 6.2% and Parsa 5.9%. Most of the districts are in the Terai region, which collectively contribute 67.2% of total manufacturing activities in the country. Hills region contribute 30.9%. Central and eastern development region together contribute two-thirds of total manufacturing activities.

In construction, Kathmandu tops the list with 15.8% contribution, followed by Lalitpur 4.6%, Morang 3.1%, Dhanusha 3%, and Jhapa 2.8%.  Hills region contribute almost 53.3% of total construction activities in the country, followed by Tarai 39.8% and mountains 6.9%. Central development region contributes 49.2%.

In retail and wholesale trade, Kathmandu tops the list with 19.9% contribution, followed by Morang 4.7%, Rupendehi 3.6%, Jhapa 3.4% and Lalitpur 3.4%. Tarai tops the list with 49.2% contribution.  Central development region contributes 46.6%.

In real estate, renting and business activities, Kathmandu tops the list with 19.4% contribution, followed by Lalitpur 5%, Chitawan 4.7%, Kaski 4% and Sunsari 3.8%. Hills region contribution is 52.4% and Tarai 43%. Central development region contributes 49.8%.

So, what is the relationship between district per capita GDP and poverty rate. Well, the chart above shows that there is an overall negative relationship, i.e. higher district per capita GDP is associated with lower district poverty rate (the red line shows a liner fit). Now, Manang and Mustang seem to be an outlier as the population there is the lowest (Manang 6,538 and Mustang 13,452), but economic activities seems to be moderately good. It is interesting that poverty rate, however, is pretty high (Manang 37% and Mustang 40%). It might have to do with the high seasonal fluctuation in economic activities (trekking season means robust value added economic activities, but heavy snowfall means a slowdown).  Construction, and hotel and restaurant activities are relatively bigger than other economic activities. [District per capita corresponds to FY2013 and poverty rate corresponds to FY2011.]

Macro messages:

1. The Tarai and Hills remain central to the economy. Tarai is particularly important for agriculture sector as a majority of the agricultural land and population live there. Being close to the Indian borders and housing major manufacturing firms, it also remains an important industrial hub. Hills region drive the services sector, which contributes the most to GDP growth and is also the largest employment provider. As an administrative hub endowed with probably the best quantity and quality of infrastructure (though still at poor standing compared to similar cities in comparator countries), Kathmandu district tops the list in its contribution to overall industry and services sectors.
2. Manang, Mustang, Humla, Mugu, and Dolpa consistently come at the last on economic activities, understandably due to the difficult terrain, harsh weather and the lack of adequate infrastructure to stimulate economic activities. Also note that the cost of physical and social infrastructures in these regions is also high.
3. Obviously, the quickest route to poverty reduction is the stimulation of local economies, thus generating gainful employment opportunities and higher per capita income. This means districts with relatively low economic activities need more attention. However, this has to be done in sync with other investment projects so that robust value chains (in all sectors) are created and districts are interlinked based on their comparative (and competitive) advantages. A sensible idea would be to promote niche products, nurture them and eventually link them to national value chains. Investments and development interventions done in silos and without a coherent, sensible strategy to link up various value added activities across the country are not going to yield the desired results, i.e. sustainable poverty reduction and employment generation. This applies to all sectors and thematic focuses.

It is pretty difficult to see the relevant stuff on the figures above. One way to see full size version is to right click on the figure, select 'copy image URL' and then paste it in browser. Also, it was not possible to include all the figures in this blog post. Below is a slideshow for interested folks.

 

Bill Gates on Jeff Sachs and international development

Bill Gates writes:

---

In the end, I hope poverty fighters will not let what they read in this book stop them from investing and taking risks. In the world of venture capital, a success rate of 30 percent is considered a great track record. In the world of international development, critics hold up every misstep as proof that aid is like throwing money down a rat hole. When you’re trying to do something as hard as fighting poverty and disease, you will never achieve anything meaningful if you’re afraid to make mistakes.

I greatly admire Sachs for putting his ideas and reputation on the line. After all, he could have a good life doing nothing more than teaching two classes a semester and pumping out armchair advice in academic journals. But that’s not his style. He rolls up his sleeves. He puts his theories into action. He drives himself as hard as anyone I know.

---

NEPAL: Major findings of Annual Household Survey 2012/13

The CBS is publishing annual household survey (AHS) data (no online link to the document yet) starting FY2013 to compliment the low frequency surveys (NLSS, NLFS). These high frequency survey data on key household characteristics could play an important role in steering public policy and development debates in the right direction. Basically, data on demography, housing, consumption and employment/economic activity are included. The household consumption data may also used to estimate final consumption figures in the national accounts estimate. Exercise caution while comparing annual household survey data with that of NLSS and NLFS. The latter ones are standard surveys that are used to compare various indicators (poverty, inequality, labor force participation, etc) within and across countries. The annual survey may not be as forceful as the periodic surveys. That said, they can still be used to capture changing expenditure and employment dynamics.

The average nominal per capita consumption in 2012/13 was NRs44,596, with the share of poorest 20% population at 7.6% and the richest 20% population at 45.8%. In urban and rural areas, per capita consumption was NRs84,134 and NRs.36,694, respectively.

Regarding consumption pattern, 59.2% of household expenditure was spent on food, followed by 11.6% in rent, 4% in alcohol and tobacco, 3% in education, 1.1% in durables, 0.8% in utilities, and 20.4% in non-food items. Looks like households spent more on alcohol and tobacco than education and durables combined! As a share of consumption, the poorest quintiles consume more alcohol and tobacco than the richest quintiles. Also, high inflation is particularly going to be severe in the case of poorest households because they spend about 67.1% of income on food and 17.2% on non-food items. For households in the richest quintile, this is 40.6% and 26.4%, respectively.

The AHS 2012/13 puts unemployment rate at 3.3% and labor force participation rate at 81.1%. Labor force participation includes the population employed (at least for an hour) in the past 7 days and those actively searching for jobs. Discouraged workers are not included in the labor force. Male and female unemployment rate stood at 3.2% and 3.4%, respectively. While urban areas had the highest unemployment rate (8%), folks between 20 and 24 years had unemployment rate of 7.7%. Also, unemployment was highest in the richest quintile (5.3%).

Labor underutilization rate stood at 27.8%, which includes unemployment rate (3.3%), time related underemployment (13.4%), skills mismatch (4.2%), and inadequate earnings (6.9%). Labor underutilization rate is highest in urban areas and also among households in the highest consumption quintiles. Those people who are willing to work for more than 40 hours in a week, but are not getting that much of work hours are clubbed under time related underemployment. Those folks whose skills are not fully utilized in the present job are clubbed under skills mismatched. Those people earning is less than half of the mean income are clubbed under inadequate earnings.

Agriculture, forestry and fishery accounted for about 66.5% of total employment in 2012/13, followed by 6.8% in wholesale and retail trade, 5.2% in manufacturing, and 3.4% in education.

A majority of the population is of working age (15-59 years) — almost 57%. There are 88.5 males per 100 females (partly reflective of the large number of males who go abroad) and average household size is 4.6.  About 15.4% of households have 1-2 persons and 37.9% have 3-4 percent, meaning that 53.5% of households have less than or equal to 4 persons. About 36.6% of households are headed by less than 40 years old. About 25.3% of households are headed by females.

While a majority of the households owns a house, just 10.2% households take it on rent for living. But, the renter households are higher in urban areas (38.8%) than rural areas (just 3.6%). In urban areas, 57.5% households own a house. Now, the richest quintile appears to rent more than the lower quintiles. A general trend is that the richer your household is, the higher the probability that you will be living in a rented place. Quite surprising!

Furthermore, while a majority of dwelling of households in rural areas is mud bonded, in urban areas the majority of dwelling is pillar bonded. Given the large number of households in rural areas, the overall average for Nepal is 49.5% dwellings that are mud bonded. Also, the poorest quintiles tend to have more mud bonded dwelling (58.7%) and the richest quintiles tend to have pillar bonded (45.3%). Sounds obvious, but it is always good to get the near-exact  numbers.

About 43.7% of households have toilet with flush connected septic tank and just 6.1% of households have toilet with flush connected to public sanitation. About 20.2% of households in the richest consumption quintile have toilet with flush connected to public sanitation. Overall, 31.6% of total households do not have toilet facility (among lowest consumption quintile households, the figures is 63.4%).

In terms of access to ICT, 82.1% of households have access to mobile phone, with households in the richest quintile have 95.4% access to mobile phone. About 60% of households in the poorest quintile have access to mobile phone.

Trip to Manang (& Pokhara)

Post by Chandan Sapkota.

Road

Road

Mini waterfall.

Lamjung Himal

Commercial sheep farming in Timang.

Dhukurpokhari, Manang, @3060 meter

Phewa Lake

Begnas Lake