Thursday, August 17, 2017

India after demonetization (growth, public finance and inflation)

A lot has been written about the impact of the demonetization shock on 8 November 2016 in India. PM Narendra Modi's government took a surprise decision to withdraw IRs1000 and IRs500 denomination currency notes (about 86% of value of total notes in circulation) to rein in corruption, counterfeiting, terrorist financing, and accumulation of black money (income that’s hidden from tax authorities). While the move is generally good for the economy over the medium and long terms (higher GDP growth, tax compliance, tax revenues, digitization, lower interest rates as liquidity in banking sector increases, etc), it had significant impact on economic activity in the short term, especially due to the botched implementation (shortages of cash, confusion, temporary withdrawal limits, sporadic bank closure due to the inability to withstand demand, etc).

Here are some charts that depict the partial impact of demonetization on economic activities. Note that others factors are also at play here and it is hard to isolate just the impact of demonetization. However, one thing is clear: this shock increased the slope of deceleration for at least two to three quarters.

Provisional estimates show that economic activities (at basic prices) decelerated to 7.1% in FY2017 (April 2016 – March 2017) from 8% in FY2016. GVA growth (at basic prices) was 6.6% in FY2017. While the favorable monsoon boosted agricultural output by 4.9%, industrial and services outputs grew by 5.6% and 7.7%, respectively. These are lower than 8.8% and 9.7%, respectively, in the previous year. Within industrial sector, mining and quarrying and construction outputs decreased the most as demonetization (mainly through shortage of cash to conduct daily transactions) dented transactions and payments (both formal and informal ones). Mining and quarrying grew by just 1.8% compared to 10.5% in FY2016. Similarly, construction activities grew by 1.7%, sharply down from 9.7% in FY2016. Manufacturing activities also slowed down compared to the previous year. Meanwhile, within services sector, wholesale and retail as well as hotel and restaurant activities slowed down. Same with financial and real estate activities. Note that, agriculture, industry and services sectors constitute 15%, 31% and 54%, respectively, of gross value added. 

If we look at the quarterly estimates of GDP, the deceleration is faster after Q2 of FY2017 (November falls in Q3). Economic activities are usually robust in the first two quarters as the government tends to front-load spending and households tend to purchase goods ahead of the festival season in the third quarter. However, in FY2017, economic activities, especially industrial and services, slowed down in all the four quarters. In a way, the economy was starting to cool off even before the shock in the third quarter and then demonetization accelerated the downward slide. Also, note that in FY2016 India updated the way GDP is computed, especially moving from GDP at factor cost to reporting GVA at basic prices plus base year was changed to 2011-12 for GDP, IIP, CPI, and WPI as well. As such, nominal GDP growth (market prices) in FY2017 is provisionally estimated at 11%, which is marginally lower than the no-demonetization scenario of about 11.9% (CSO’s estimate before demonetization).

Now, the expenditure side data reveal kind of similar narrative. Consumption grew by 10.5% (compared to 5.7% in FY2016) led by a notable growth of government consumption (20.8% vs. 3.3% in the previous year). Private consumption, which constitutes about 84% of total consumption and 56% of total GDP at market prices, grew slightly higher than in FY2016. Demonetization seems to have dented investment the most, pulling down overall GDP growth. Gross capital formation grew by just 1.7%. Within it gross fixed capital formation grew by 2.4% (compared to 6.5% in the previous year), led by a dip in private fixed capital formation across all sub-sectors (particularly investment in dwellings and building structures, and machinery and equipment— these together account for about 91% of private GFCF). 

There was some improvement in net exports (as export growth outstripped import growth), but it was not sufficient to compensate for the dip in investment. The big gap between the GDP growth (market prices) line and the stacked up bars in FY2016 is due to the large discrepancies, which shot up drastically in that particular year (remember the data revision mentioned above and the statistical discrepancies associated with it).

If you look at the quarterly GDP data on expenditure side, the narrative is similar to the one mentioned above for quarterly GDP data on supply side. After the demonetization, government consumption increased drastically but gross investment and net exports slowed down. In fact, gross capital formation and net exports registered negative growth in Q4. 

Now, let us look at the impact on public finance. The rush to declare hidden income and deposit it in banks after November 2016 seems to have favored the government (increase in revenue) and banking sector (increase in liquidity). Revenue (tax and non-tax) increased by 19.1%, the highest in the last six years (surprisingly, share of monthly revenue between Nov-Mar remained pretty much unchanged: 49.5% in FY2017 vs. 50.8% in FY2016— folks definitely paid more taxes in the last five months of FY2017 compared to FY2016, but the distribution was same).

As a share of GDP, center’s revenue increased to an estimated 9.4% of GDP (quite low considering the revenue mobilized by other comparable emerging economics, but then we need to also look at combined revenue of center and state, which is around 20% of GDP), up from 8.7% of GDP in the previous year. Recurrent spending grew by 12.8%, up from 4.8% (this complemented to some extent the slowdown in private consumption) but capital spending growth slowed down to 10.6% from 28.6% in the previous year. Overall, fiscal deficit is estimated to be within 3.5% of GDP and primary deficit 0.3% of GDP (both an improvement compared to previous year). 

Demonetization seems to have had a favorable effect (together with bountiful harvest due to good monsoon, and low fuel prices among others) on inflation by dampening consumer demand. Inflation has been decelerating in recent months and is far below 6.1% recorded in July 2016. Food prices, which has 45.9% weight in CPI basket, have cooled off substantially, but consumer prices of tobacco have grown pretty consistently, and housing and fuel prices are heating up. Demand for durables and everyday items (under miscellaneous heading) is cooling off. Overall, annual average (monthly) inflation of 4.5% for FY2017 is between RBI’s lower and upper bounds of 2% and 6%, respectively (average of 4%). The largest contribution to CPI inflation in FY2017 came from food & beverage prices, followed by miscellaneous items; housing; clothing and footwear; fuel and lighting; and pan and tobacco products. 

The wholesale price index shows a bit different picture. It was on a downward spiral since July 2104 until August 2015, and then started to pick up till February 2017 before declining. The WPI inflation for three months following the demonetization shock shows no sign of modereation. However, after that it seems to have had an impact. 

If we look at IIP, then there seems to be an impact after November as demand for manufacturing goods (which has 77.6% weight in IIP basket) moderated substantially. Specifically, demand for primary, and capital goods, and consumer durables slumped (in other words, contraction of urban demand). In May, IIP growth was negative. This might have come from both demand and supply sides. The depressed consumer demand may be caused first by the immediate drop in formal sector demand and then by a (second round) drop in informal sector demand (official statistics pick up the first one faster and then the slowdown in formal sector is reflected as further dip in formal sector demand after few months). In other words, the demand shock in informal sector takes some time to reflect as demand depression in formal sector. This might have slowed down supply of investment. Meanwhile, demand for investment may also be slowing down (and consequently bank credit has contracted) due to waning confidence about returns and consumer demand. 

Anyway, both CPI and WPI inflation are expected to rise (but still stay between the RBI’s lower and upper bounds) as demand depressing effect of demonetization dissipate and consumer demand for both perishable as well as durable goods gain some momentum (especially coming from the hike in public salary and allowance, and heating up of vegetable prices). Also, GST may lubricate the rusty industrial engines and boost both production and investment. Government, meanwhile, needs to enhance capital spending by accelerating project implementation (stuck by a myriad of bureaucratic as well as legislative factors— land, environment, tax incentives clearance, etc). Both private and public capital formation needs to increase. It needs to be seen how overly leveraged companies will increase investment and how government speeds up capital spending (especially in rural housing, roads & bridges, and other infrastructure projects; but potential farm loan waivers by states mean that they will have less fiscal space to increase capex).

GVA growth (at basic prices) is targeted at 7.3% for FY2018.

Wednesday, August 9, 2017

Nepal's central bank needs younger talents

It was published in The Kathmandu Post, 07 August 2017

Nepal Rastra Bank needs to be objective and pragmatic to maintain institutional credibility

The Nepal Rastra Bank (NRB) should ideally operate without courting much controversy over its internal management, operational modality, assessment of the economy, and the regulatory and supervisory mandate. Staying above reproach would ensure financial stability as well as the achievement of its two main macroeconomic targets—curbing inflation and boosting the growth rate. NRB needs to be as objective and pragmatic as possible to maintain institutional credibility and integrity on the execution of monetary policy.

Unfortunately, this supposedly impartial autonomous institution is in the news these days for all the bad reasons; including the leadership’s ill-intentioned plan to do away with the 30-year employment limit, ad hoc changes to monetary rules and regulations, and weak monitoring and evaluation of banks and financial institutions (BFIs).

One of the most controversial issues faced by the central bank currently is internal lobbying to do away with the 30-year employment limit for its employees. This has been a recurring issue since 2001, when such a provision was introduced under the financial sector reform project. Currently, employees are mandated to retire if they have served for 30 years or are above 58 years of age. NRB recently formed a new committee to suggest amendments to its employment regulations. Many suspect this to be a pretext to end the threshold on the years of service. This has generated mixed reaction within and outside of NRB. While some have expressed strong reservation over the intention to form the committee, others either have remained silent or are actively lobbying to end the limit on years of employment. 

There is no justifiable logic in changing current terms and conditions for employment. Some argue that ending the age limit on years of employment will ensure that a large proportion of senior employees will be able to retain their jobs for the time being, avoiding disruption to service delivery caused by retirement of old employees and loss of institutional capacities.

This is hogwash, as the NRB needs young talent in all its departments to effectively execute its policies, undertake research activities, and fulfil its regulatory and supervisory mandate. The marginal benefit of retaining old employees is far less than the marginal cost of hiring young people and training them to take on the required job responsibilities.  The central bank needs to attract top talent from domestic as well as foreign universities by offering attractive entry-level research or management positions. Importantly, a fresh breath of air is needed at NRB so that its actions are proactive rather than reactive. The latter has been the case for the past few years as NRB has repeatedly missed clear signals of disruptive credit flows and accumulation of unbalanced portfolio by BFIs.

NRB’s gain as well as the gain for the entire financial sector will be higher if the old guns follow the current employment regulation and make way for younger talent. The constant tussle among its old and new employees, politically-affiliated unions, and rolling out ad hoc management regulations undermines NRB’s credibility and distracts it from focusing on its main tasks: helping the government achieve growth and inflation targets without jeopardising financial stability, enhancing access to finance, and effectively supervising and regulating the financial sector.

Drifting away

The other important issues that are bogging down NRB are changes to set monetary rules and regulations, and weak monitoring and evaluation of BFIs. First, responding to the sudden credit squeeze—a result of the BFIs repetitive practice to accelerate credit growth more than deposit growth—NRB became overly accommodative by tweaking accounting rules in computing the credit to core capital-cum-deposit (CCD) ratio in its mid-year review of the fiscal year (FY) 2017 monetary policy. This ad hoc policy change was in effect a reward for the BFIs struggling to mend their ways after recklessly increasing credit to few unproductive sectors.

NRB allowed BFIs to discount 50 percent of productive lending, plus lending to deprived sectors and the agro sector at subsidised interest rates, while computing the CCD ratio. It gave some breathing space to BFIs to rework on their lending practices and meet the mandatory threshold of 80. However, NRB rolled back this provision in its FY2018 monetary policy. This kind of overly accommodative ad hoc measure has fostered a moral hazard, whereby BFIs continue to act recklessly , anticipating a reprieve from NRB in case of negative consequences. In doing so, they privatise profits and socialise losses. NRB has acted on behalf of the BFIs rather than acting as their supervisor and regulator, and committing to protect people’s deposits and return on savings. 

Second, the central bank failed to stabilise interest rates, which have remained volatile in recent years. Retail interest rates have gone up sharply, a result of the reactive working culture at NRB. Its monetary instruments to smooth liquidity flows have so far been ineffective. Recently, it changed its approach to maintaining retail interest rates between 3 and 7 percent. As long as NRB uses selling and buying of treasury bills and bonds as its main instrument to manage liquidity and monetary policy, retail interest rates will continue remain volatile. It’s a management problem as well as being a demand-supply problem. NRB needs to be more creative in liquidity management, and monitoring and evaluation of BFIs.

Lastly, in its monetary policy for FY2018 the central bank is aimlessly throwing arrows in all directions to achieve economic growth and inflation targets of 7.2 percent and 7 percent, respectively. NRB is yet to explain how it is going to achieve a higher growth rate than in FY2017, which itself was largely a ‘base effect’ blessing, by virtually keeping unchanged money supply growth and policy rates. Liquidity will likely be tight soon owing to the continuing deceleration of remittance income and the slow as well as uneven pattern of government spending

Meanwhile, a bloated budget dependent on large deficit financing is going to put upward pressure on interest rates. This is going to increase borrowing costs for businesses and individuals and eventually slow down economic activities, especially if the government is unable to spend the money it collects from taxes and borrowing. Given past experience, the FY2018 budget will likely be under spent. 

Orderly house

NRB needs to keep its house in order so as to be an effective regulator of the financial sector, a catalyst for enhanced financial inclusion, and a vibrant knowledge hub. Specifically, it needs to set an example by not tinkering its employment regulations to benefit old employees. Instead, it should allow young talent to take up challenging roles, thus fostering creativity, policy innovation and better service delivery.

It should also desist from rolling out ad hoc policy measures to reward moral hazard behaviour in the financial sector. A stop-gap, ad hoc policy as well as management style undermines its credibility and trust.

Sunday, August 6, 2017

Why does the government under-execute capital budget?

Towards the last quarter of fiscal year, pretty much everyone talks about slow capital spending. Parliamentary committees issue notices/instructions to government agencies to accelerate spending. Ministry of Finance summons line ministries’ representatives to review progress on stated milestones every quarter. The media also gives ample space to spending level and pattern, raising questions over hasty spending on shoddy works before the end of fiscal year. The same reasons are debated and recycled during each quarterly review meeting. The cycle of issuing notices, review of quarterly portfolios and pondering over the same reasons for slow capital spending keeps on rotating each year. Unfortunately, nothing substantial comes out of it (with respect to achieved outputs).

The government and MOF said FY2017 (mid-July 2016 to mid-July 2017) would be different because budget was announced one-and-a-half months prior to the start of the fiscal year. They said it will give ministries adequate time to get approval for spending and initiate preparatory project planning (especially procurement documents), all of which were expected to accelerate capital spending. Specifically, the idea was to finish all preparatory work and start issuing tender notices and in some cases finalize contractors before the start of festival season (September-November). 

So, was the result? Pretty much the same or even worse: 
  • First, the spending pattern (i.e. quality of spending) hardly changed despite the early approval of budget. Almost 60% of the actual capital spending happened in the last quarter and 41.2% in the last month. It raises doubt over the quality of spending. Often, spending (and approval of payments for completed as well as some pending works) is accelerated in the last month by doing shoddy work. The haste in spending without any quality control by the ministries (what is the monitoring and evaluation unit doing?) leads to cheapjack construction. This in turn increases operation and maintenance budget (which is a part of recurrent spending) for the next few years. 
  • Second, capital spending absorption capacity has receded. Just 65.5% of planned capital budget was spent in FY2017 (NRs204.3 billion spent vs NRs311.9 billion planned). Although it is slight higher than 58.6% in FY2016 (when the economy was crippled by trade and supplies disruptions), it is still lower than 76% in FY2015 (when earthquakes struck around the last quarter of fiscal year). As a share of GDP, capital spending increased to 7.9% of GDP, up from 5.4% of GDP. The planned capital spending in FY2017 was 12% of GDP.
Now, the MOF sometimes points to the National Reconstruction Authority (NRA) for slow capital spending as reconstruction work did not pick up steam as expected. However, NRA itself was burdened with approval from Cabinet, lack of cooperation from line ministries, and the hooks kept by the MOF on NRA’s discretion in spending the allocated budget (including to cover procedural administrative costs). These exerted an inertia on NRA’s speed and efficiency. Nevertheless, preliminary data from the MOF’s ‘Red Book’ itself show that NRA did quite well in FY2017. It was asked to spend around NRs140 billion in housing grants, reconstruction, and administrative expenses. It seems to have kept the promise for now as per preliminary data. We will have to look at the full year data to see how close NRA was to the spending target. The pick up in reconstruction activities boosted growth of construction, real estate and housing, and mining & quarrying activities in FY2017. In FY2018, NRA is asked to spend NRs146.2 billion on reconstruction of infrastructure and housing grants. 

So, it seems the same factors have been constraining capital spending and approving budget early is not going to change that unwanted steady equilibrium drastically. Budget execution is affected by: (i) structural weaknesses in project preparation and implementation (i.e. lack of initial planning); (ii) low project readiness (no feasibility studies, land acquisition, environment clearance, detail design and procurement milestones); (iii) bureaucratic hassle in project approvals and sanctioning of spending; (iv) weak project and contract management (high staff turnover and inability to rein in errant contractors); and (v) political interference at planning, management and operational stages. FY2017 budget spending was also affected by local elections as government staff were deputed to conducting and monitoring elections instead of managing projects. 

These key issues that slow down spending have more to do with the capacity of government staff and the strength of their offices (institutional memory, motivation, etc) than early release of budget. Of course, early approval of budget provides certainty of spending but it does not directly address the above mentioned issues. 

For project readiness, the NPC should be assisting line ministries and local bodies to conduct pre-feasibility studies. Importantly, it should be appraising the project proposals prepared by line ministries and local bodies keeping in mind factors such as land acquisition, rate of return, cost-benefit analysis, implementation modality, etc. These then should be aligned with medium term expenditure/revenue framework. The NPC should be playing a pro-active role in assisting line ministries and local bodies in project planning and appraising (a sort of localized ‘project bank’). 

Meanwhile, the MOF should be keeping tab on expenditure allocation, especially restraining the temptation to allocate budget for projects that are not ready or are not sanctioned by the NPC. This usually happens because of political pressure. Furthermore, MOF should closely collaborate with NPC and OPMCM (especially their M&E units) to monitor progress and unwind constraints faced during the implementation phase (including speedy resolution of constraints by taking the issues up the political chain and if necessary the cabinet). The obsession with meeting revenue targets has meant that these crucial tasks are delegated as ceremonial undertakings. The MOF and NPC need to be more agile, responsive and cooperative.

Then comes the line ministries. These are primarily responsible for project conception, design and execution. The onus of faster and quality spending lies in project directors (mostly those at joint secretary level). If the government staff at project offices are lethargic, then no project will see accelerated spending. 

The role of multilateral and bilateral donors, who implement projects through the government mechanism, is also crucial as sometimes approvals or endorsements or disbursements are stuck at their offices. Again, being proactive in smart project and contract management is the key to accelerating capital spending. The hassle created by Acts and policies are not going to be solved overnight. Low hanging fruits should be harvested first.

Same applies to the various parliamentary committees that are increasingly infringing on project management (by ordering halt to procurement process, summoning multiple times for hearing but without any tangible outcome, etc) and issuing orders after orders for faster capital spending. How about these committees also give some attention to the need for project readiness, the resources needed for it, and monitoring and evaluation (in collaboration with NPC, MOF, OPMCM and local experts)? Negligence by politically-affiliated local contractors (who bid for projects that are beyond their financial and management capacity) is costing the taxpayers dearly. A strong M&E system is needed in addition to enhancing capacity of project offices to design, appraise and implement projects. 

Overall, the government met the revenue target (more on this in later blog posts), but failed to accelerate capital spending as expected. What will be the fate of FY2018 budget? Note that accelerated capital spending is one of the pillars for rapid structural transformation given Nepal's stage of development right now.