Prepared by TERI for research and strategic-information purposes; it is not investment, legal, tax or financial advice. The data cover customers' dealings with commercial banks over one week. "Purchases" are read as dollars banks bought from customers (dollar supply) and "sales" as dollars banks sold to customers (dollar demand). The report does not explain sector moves, so the drivers on this page are hypotheses ranked by how well the data fit them.
00 — OverviewExecutive Summary: A Flat Total Hiding a Big Change in Who Supplies Dollars
Retail FX covers dollars that commercial banks buy from and sell to their customers. A flat total says little; the sector mix says a lot. Last week Mining was the second-largest source of dollars to banks (USD 65.77 million of purchases, 18.1% of supply). This week it was fourth (USD 26.08 million, 8.6%). Agriculture, Tourism and Trade filled much of the gap. On the demand side, Oil & Energy increased its draw to USD 69.06 million, now 25.3% of all dollars banks sold, and Manufacturing rose to USD 37.24 million. The net effect was a retail dollar surplus (purchases minus sales) that fell from USD 121.35 million to USD 29.70 million.
What Moved Total Turnover: Contribution by Sector
Mining vs Everything Else
Key findings
- Mining is the story of the week. Its turnover fell by USD 38.34 million, more than the entire USD 29.21 million fall in the market. Almost all of it was banks buying fewer dollars from mining customers (−USD 39.69 million, −60.3%).
- Oil & Energy is the main gainer on the demand side. Turnover rose USD 15.77 million; banks' dollar sales to the sector rose USD 12.00 million (+21.0%). It is the market's biggest net user of dollars (net −USD 57.57 million).
- Construction (machinery and equipment) also fell sharply, from USD 39.17 million to USD 23.42 million (−40.2%), the second-largest drop after Mining and easy to overlook.
- Trade and Agriculture hold the base. Trade is still the largest sector at USD 172.34 million (30%), but it flipped from net supplier (+USD 13.84 million) to slight net user (−USD 4.04 million). Agriculture's net supply stayed near USD 50 million and its share of dollar supply rose from 17.3% to 21.2%.
- The dollar cushion thinned but did not vanish. The retail market is still in net surplus (+USD 29.70 million). Illustratively it would take about 11% more dollar sales, about 10% fewer purchases, or a 43% rise in oil and energy sales alone to erase it.
- The shilling has barely reacted: TZS 2,636.74 per dollar, +0.14%. The risk is persistence: weak mining supply and firm oil and manufacturing demand repeating for several weeks.
The retail dollar market is more dependent on fewer sectors than it looks: Trade, Agriculture and Tourism now carry the supply side while Oil & Energy and Manufacturing carry the demand side. Mining's swing, large and lumpy, can move the whole market's balance in a week. This week the shilling absorbed it. The next tabs explain the sector map, the likely drivers, and how big a shock would need to be to push the shilling.
01 — The Sector MapWho Supplies Dollars, Who Uses Them, and How the Shares Moved
| Sector | Previous week (USD m) | This week (USD m) | Change in turnover, USD m | Change, % | ||||
|---|---|---|---|---|---|---|---|---|
| Purch. | Sales | Total | Purch. | Sales | Total | |||
| Trade | 97.56 | 83.72 | 181.28 | 84.15 | 88.19 | 172.34 | −8.94 | −4.9 |
| Oil & Energy | 7.73 | 57.06 | 64.78 | 11.49 | 69.06 | 80.55 | +15.77 | +24.3 |
| Agriculture | 62.74 | 10.27 | 73.01 | 64.28 | 13.81 | 78.09 | +5.08 | +7.0 |
| Manufacturing | 10.29 | 27.99 | 38.28 | 13.90 | 37.24 | 51.15 | +12.87 | +33.6 |
| Individual retail customers | 25.38 | 15.84 | 41.22 | 18.82 | 17.20 | 36.02 | −5.20 | −12.6 |
| Mining | 65.77 | 8.41 | 74.18 | 26.08 | 9.76 | 35.84 | −38.34 | −51.7 |
| Tourism | 24.81 | 2.07 | 26.88 | 28.55 | 3.92 | 32.48 | +5.60 | +20.8 |
| Transport & Communication | 29.80 | 6.03 | 35.84 | 21.16 | 10.12 | 31.29 | −4.55 | −12.7 |
| Construction (mach. & equip.) | 19.44 | 19.73 | 39.17 | 9.92 | 13.50 | 23.42 | −15.75 | −40.2 |
| Social services | 10.44 | 5.31 | 15.75 | 11.58 | 5.56 | 17.15 | +1.40 | +8.9 |
| NGOs | 9.05 | 2.37 | 11.42 | 12.01 | 0.42 | 12.43 | +1.01 | +8.8 |
| Breweries / Beverages | 0.13 | 2.96 | 3.09 | 0.75 | 4.19 | 4.93 | +1.84 | +59.5 |
| Total | 363.12 | 241.77 | 604.89 | 302.69 | 272.99 | 575.68 | −29.21 | −4.8 |
Table 1.1. Weekly commercial banks' retail FX by sector. Source: Bank of Tanzania, Table 2; changes and percentages are TICGL/TERI calculations.
Where Banks Get Their Dollars: Share of Purchases
Who Uses Banks' Dollars: Share of Sales
Change in Supply vs Change in Demand, by Sector
Net Dollar Direction as a Share of Sector Turnover
The three biggest sectors by turnover (Trade, Oil & Energy, Agriculture) made up 57.5% of the week, up from 54.3% for the top three the week before (Trade, Mining, Agriculture). On the supply side, Trade, Agriculture and Tourism provided 58.5% of dollars bought this week, against 51.0% a week earlier, so the supply base became more reliant on fewer, steadier sectors after mining pulled back.
02 — Mining, Oil & the RestReading the Sectors One by One
Mining: Purchases, Sales and Net
Oil & Energy: Purchases, Sales and Net
Mining's net inflow to banks fell from USD 57.36 million to USD 16.32 million. Mining firms convert export receipts in batches, so a single week can swing sharply for reasons of timing: shipment dates, when proceeds are repatriated, and how long firms hold dollar balances. Mining remained a net supplier, but at 45% of its turnover versus 77% a week earlier. In shilling terms, the USD 41.04 million drop in mining's net position is about TZS 108 billion at TZS 2,636.74.
Oil and energy buys dollars almost one way: sales to the sector were 86% of its turnover this week. The USD 69.06 million bought is about 25% of all dollars banks sold. Because fuel imports are priced in dollars and needed regardless of the exchange rate, this demand is hard to defer, so it tends to rise when supply thins. The report gives no volume or price breakdown, so it cannot separate more fuel imported from higher fuel prices.
Trade and Agriculture: The Anchors
Construction, Manufacturing and Transport
| Sector | What changed | Why it matters |
|---|---|---|
| Trade | Purchases −USD 13.41m, sales +USD 4.47m; net +13.84 to −4.04 | Largest sector and the biggest swing in net after Mining; importers and exporters now offset each other. |
| Agriculture | Purchases +USD 1.54m to 64.28m; sales +3.54m | Stable dollar supplier, now 21.2% of supply. |
| Tourism | Purchases +USD 3.74m to 28.55m (+15.1%) | Third-largest supplier by share of purchases (9.4%), ahead of Mining this week. |
| Manufacturing | Sales +USD 9.25m (+33.1%) to 37.24m | Second-largest net user of dollars after Oil & Energy; ties to input and equipment imports. |
| Transport & Communication | Purchases −USD 8.64m; net +23.77 to +11.04 | A quiet source of supply that halved. |
| Construction | Purchases −9.52m, sales −6.23m; total −40.2% | Both sides fell; may reflect project or equipment-payment timing. |
| Individual retail customers | Purchases −6.56m; net +9.54 to +1.62 | Household supply of dollars nearly disappeared as net. |
Table 2.1. TICGL/TERI reading of Table 2 of the Bank's weekly report.
03 — What Drives the ShiftsHypotheses, Evidence and What Would Confirm Them
The Bank's report publishes the flows, not the reasons. The table below sets out the most plausible drivers for each major shift, what in the data supports them, and what to look for next week. "Fit" rates how well the published data support each driver, not the probability that it is true.
| Shift | Candidate driver | What the data show | What would confirm it | Fit |
|---|---|---|---|---|
| Mining purchases −60% | Export-receipt timing | Mining sales hardly changed (+USD 1.35m); only purchases fell, as if a batch of receipts was converted the week before. | Mining purchases back toward USD 40–65m within 1–2 weeks | Strong |
| Firms holding dollars longer | A fall in sales to the bank with no matching rise elsewhere is consistent with holding balances; not visible in this report. | Foreign-currency deposit data in the monthly review | Possible | |
| Lower mineral export volumes or prices | One week cannot show a price or volume trend. | Export values in the next monthly review | Not testable here | |
| Oil & Energy sales +21% | Import settlement and month-end payments | Sales rose through a week ending near month-end; the demand is concentrated in one sector. | Sales stay above USD 65m into November | Strong |
| Higher import bill or stock rebuilding | Cannot separate volume from price in the data. | Fuel import data; pump-price announcements | Possible | |
| Manufacturing sales +33% | Input and equipment imports | Sales +USD 9.25m, purchases +USD 3.61m; classic user profile. | Persistence over two weeks | Strong |
| Trade net flips to a user | Importers' payments rising while exporters' receipts fall | Purchases −USD 13.41m; sales +USD 4.47m. | Trade net stays negative | Partial |
| Construction −40% | Project payment timing | Both purchases and sales fell, which points to fewer transactions and not a one-way move. | Rebound next week | Partial |
| Agriculture and Tourism steady or up | Seasonal export and visitor inflows | Agriculture purchases +USD 1.54m; Tourism +USD 3.74m. Seasonal patterns suggest autumn is within the main crop-marketing and early high-season period (general context, not shown in the report). | Continued purchases in coming weeks | Consistent |
Table 3.1. TICGL/TERI assessment from one week of data. Seasonal remarks are general context and should be verified against export and arrivals data.
Supply Concentration: Share of Dollar Purchases From the Three Biggest Suppliers
The week's pattern shows an asymmetry that matters for the shilling: supply from export sectors (especially Mining) can drop by half in a week, while demand from fuel, manufacturing inputs and trade rises or holds. When both happen at once the net surplus falls quickly; when supply returns it can rebuild just as fast, as last week's USD 121 million shows.
04 — Impact on the ShillingFrom Sector Flows to the Exchange Rate: Channels, Size and Shock Arithmetic
The shilling closed the week at TZS 2,636.74 per US dollar, a 0.14% depreciation from TZS 2,633.12 (TZS 3.62). That is small. The question is whether the sector shifts could turn into something larger, so this tab traces the channels and sizes the shocks.
4.1 What happened this week
- The shilling depreciated 0.14% against the dollar. Sterling and yen moves were larger, but those mostly reflect those currencies' own moves against the dollar.
- The retail market stayed in net surplus (+USD 29.70 million), so banks did not need to buy dollars from the central bank to meet customer demand.
- The interbank cash rate stayed within 6.16%–6.59% every day, against a policy rate of 6.25% and a corridor of 4.75%–7.75%, so shilling liquidity did not tighten.
- IFEM turnover rose to USD 60.80 million: the visible footprint of banks rebalancing, covered in our companion IFEM analysis.
4.2 How big a shock would erase the surplus?
The surplus of USD 29.70 million is the cushion between dollars coming in and dollars going out through banks. The table asks how large a single move would have to be to remove it, holding everything else equal. These are arithmetic illustrations, not forecasts.
| Single change | Base this week (USD m) | Change needed (USD m) | As % of base |
|---|---|---|---|
| Total dollar sales rise | 272.99 | +29.70 | +10.9% |
| Total dollar purchases fall | 302.69 | −29.70 | −9.8% |
| Oil & Energy sales rise | 69.06 | +29.70 | +43.0% |
| Manufacturing sales rise | 37.24 | +29.70 | +79.8% |
| Trade purchases fall | 84.15 | −29.70 | −35.3% |
| Agriculture purchases fall | 64.28 | −29.70 | −46.2% |
| Mining purchases fall to zero | 26.08 | −26.08 (maximum) | −100% is not enough |
Table 4.1. TICGL/TERI arithmetic on the Bank's Table 2 totals. Mining alone can no longer erase the surplus because most of its supply has already gone; the risk now comes from broad-based demand growth or a fall in the steadier suppliers.
Size of Change Needed to Erase the Net Surplus
Net Surplus: Three Illustrative Cases
4.3 The shilling value of the sector swings
Sector Swing in Net Dollar Position, in TZS Billion
The USD 3.62 rise in the dollar price cost customers who bought dollars from banks roughly TZS 1.0 billion more across the week's USD 272.99 million of sales (about TZS 0.25 billion for Oil & Energy's USD 69.06 million). The number is small; its value is in showing that for sectors that buy about 25% of the market's dollars, each 1% move in the rate adds about TZS 1.8 billion a week to the fuel import bill at the current volume (USD 69.06m × TZS 26.4).
Oil and energy and manufacturing buy dollars to import goods whose shilling cost passes into fuel, transport and input prices. The sectors that supply dollars (Mining, Agriculture, Tourism) benefit when the shilling weakens. A depreciation therefore moves income between these groups as well as affecting prices.
Shock calculator: what happens to the net surplus?
Starting point is this week (net +USD 29.70m). Enter percent changes to selected flows; the page shows the new net position and what it is worth in TZS billion at TZS 2,636.74. Mining purchases cannot fall below zero.
05 — Outlook & TICGLWhat to Watch, and What It Means for Your Business
5.1 Three paths for the next few weeks
The paths below are illustrative and rest on one week of data. No probabilities are attached.
| Path | Mining purchases | Oil, Manufacturing, Trade demand | Net surplus and shilling |
|---|---|---|---|
| A. Mining rebounds | Back toward USD 40–65m | Stable | Net surplus rebuilds above USD 30m; shilling steady |
| B. Mining stays low, demand firm | Stays near USD 26m | Oil sales stay near USD 70m; manufacturing near USD 37m | Net surplus thin but positive; mild shilling weakness; IFEM stays active |
| C. Broad demand squeeze | Stays low | Sales rise 10% or more across oil, manufacturing and trade; steady suppliers (Agriculture, Tourism) soften | Net surplus turns negative; faster depreciation; more official operations likely |
Table 5.1. Illustrative paths, TICGL/TERI.
5.2 Weekly indicators to track
| Indicator | This week | Signal to watch |
|---|---|---|
| Mining purchases | USD 26.08m | Rebound toward USD 65m, or a second weak week |
| Oil & Energy sales | USD 69.06m | Above USD 70m again, or back below USD 60m |
| Net retail surplus | USD 29.70m | Falls below zero |
| Share of supply from Trade, Agriculture, Tourism | 58.5% | Falls if steady suppliers soften |
| Sales ÷ purchases | 0.90 | Approaches or exceeds 1.0 |
| USD/TZS weekly change | +0.14% | Persistent moves above 0.5% a week |
Importers
- Oil, manufacturing and trade importers compete for the same dollars.
- Plan settlement dates around month-end and the sector's rising demand.
Exporters
- Timing of conversion matters: mining purchases swung USD 40m in a week.
- Agriculture and Tourism supply is steadier.
Banks and treasuries
- Track sector net positions, not just totals.
- Interbank volumes will follow retail surplus swings.
Project sponsors and lenders
- Stress-test FX assumptions in models against a thinner dollar cushion.
- Match revenue and cost currencies where possible.
Talk to TICGL/TERI
Request a briefing on sector FX flows for your business, or a scoping call for a feasibility, market-entry or market-research study with FX and pricing scenarios.
What We Do →TICGL Markets →TERI Research →Book a consultation →✉️ economist@ticgl.comThe shilling did not move this week, but the dollar cushion did. Watch Mining's purchases and Oil & Energy's sales: they are the two numbers that decide whether a thin surplus stays positive.
TICGL/TERI reading of the Bank of Tanzania weekly data, October 202606 - Sources & AssumptionsMethod, Assumptions and Data Notes
Source report. Bank of Tanzania, Domestic Financial Markets Weekly Report, 19–23 October 2026 (Table 1, Table 2, Sections 1.1–1.3, 2.1). Calculations. Shares, changes, net positions, contributions, shock sizes and shilling conversions are TICGL/TERI arithmetic on the Bank's published figures. Limitations. One week; sector totals are customers' dealings with commercial banks; the report does not state reasons for sector moves.
Assumptions and notes
| Item | Treatment on this page |
|---|---|
| Purchases and sales | Read as banks' purchases of dollars from customers (supply) and sales of dollars to customers (demand). Net = purchases − sales. |
| Totals and rounding | Sector totals follow the Bank's published column; some differ from purchases + sales by 0.01 due to rounding. |
| Shilling conversions | USD amounts × TZS 2,636.74 (23 October indicative mean rate). |
| Ex-mining turnover | Total minus Mining: USD 604.89m − 74.18m = 530.71m last week; USD 575.68m − 35.84m = 539.84m this week. |
| Supply shares | Trade, Agriculture and Tourism: (97.56 + 62.74 + 24.81) ÷ 363.12 = 51.0% last week; (84.15 + 64.28 + 28.55) ÷ 302.69 = 58.5% this week. |
| Oil and fuel import sensitivity | Illustrative: USD 69.06m of weekly dollar sales × 1% of TZS 2,636.74 = TZS 1.82bn. It assumes unchanged volumes. |
| Seasonal remarks | General context only; not in the Bank's report. |
- Bank of Tanzania, Domestic Financial Markets Weekly Report, 19–23 October 2026.
- TICGL/TERI analysis and calculations on the above.
07 - Quick AnswersFrequently Asked Questions
How did Tanzania's retail FX turnover change in the week to 23 October 2026?
Total retail foreign exchange turnover fell 4.8 percent to USD 575.68 million from USD 604.89 million. Excluding mining, turnover actually rose about 1.7 percent, from USD 530.71 million to USD 539.84 million.
How much did mining turnover fall in retail FX?
Mining turnover fell 51.7 percent, from USD 74.18 million to USD 35.84 million. The whole fall came from bank purchases of dollars from mining customers, which dropped from USD 65.77 million to USD 26.08 million.
Why did oil and energy FX turnover rise?
Oil and energy turnover rose 24.3 percent to USD 80.55 million, mostly because banks sold USD 69.06 million to the sector, up from USD 57.06 million. The report does not give the reason; likely candidates are import settlement timing and month-end payments.
Is the retail market still in net dollar surplus?
Yes. Banks bought USD 302.69 million from customers and sold USD 272.99 million, a net surplus of USD 29.70 million, down 75.5 percent from USD 121.35 million the week before.
What would it take to erase the net dollar surplus?
Illustratively, a rise of about 11 percent in total dollar sales, a fall of about 10 percent in total purchases, or a rise of about 43 percent in oil and energy sales alone. Even if mining purchases fell to zero the surplus would not disappear.
What did the sector shifts mean for the shilling?
So far very little. The shilling closed at TZS 2,636.74 per dollar, a 0.14 percent depreciation. The risk is not this week's move but whether weak mining supply and firm oil and manufacturing demand persist.
Muhtasari kwa Kiswahili
Jumla ya Retail FX ilipungua kidogo, lakini sekta zimebadilika sana. Katika wiki ya tarehe 19 hadi 23 Oktoba 2026, mauzo na ununuzi wa dola kati ya benki za biashara na wateja ulishuka kutoka USD milioni 604.89 hadi 575.68 (asilimia 4.8). Ukiondoa sekta ya Madini, soko liliongezeka kwa takriban asilimia 1.7. Mabadiliko ya sekta ndiyo habari halisi.
Madini (Mining) yalishuka sana: kiasi cha jumla kilishuka kutoka USD milioni 74.18 hadi 35.84 (asilimia 51.7). Sababu ni benki kununua dola chache kutoka kwa wateja wa madini: kutoka USD milioni 65.77 hadi 26.08 (asilimia 60.3). Mauzo ya dola kwa sekta hii hayakubadilika karibu hata kidogo. Anguko la Madini peke yake (USD milioni 38.34) ni kubwa kuliko anguko lote la soko (29.21).
Mafuta na Nishati (Oil & Energy) yaliongezeka: kutoka USD milioni 64.78 hadi 80.55 (asilimia 24.3). Benki ziliuza dola nyingi zaidi kwa sekta hii (USD milioni 57.06 hadi 69.06), ikiwa sasa karibu robo ya dola zote zinazouzwa. Viwanda (Manufacturing) pia viliongezeka asilimia 33.6, na Ujenzi (Construction) ulishuka asilimia 40.2.
Biashara (Trade) na Kilimo (Agriculture) bado ni nguzo: Biashara ni sekta kubwa zaidi (USD milioni 172.34), lakini ilibadilika kutoka kuwa msambazaji halisi wa dola (+13.84) hadi mtumiaji mdogo (−4.04). Kilimo kinaendelea kusambaza takriban USD milioni 50 kwa jumla halisi, na sehemu yake ya ugavi wa dola imepanda kutoka asilimia 17.3 hadi 21.2.
Nini kinaendesha mabadiliko haya? Ripoti haitoi sababu, kwa hiyo haya ni dhana za TICGL/TERI: (1) muda wa kupokea mapato ya mauzo ya nje kwa Madini; (2) malipo ya uagizaji wa mafuta na ya mwisho wa mwezi; (3) uagizaji wa malighafi na vifaa kwa viwanda; (4) msimu wa mauzo ya Kilimo na Utalii. Ugavi wa dola kutoka sekta za mauzo ya nje hubadilika-badilika, wakati mahitaji ya mafuta na malighafi ni magumu kuahirishwa.
Athari kwa shilingi: Ziada halisi ya dola ilishuka kutoka USD milioni 121.35 hadi 29.70 (asilimia 75.5), lakini bado ni chanya. Shilingi ilifunga TZS 2,636.74 kwa dola, ikipungua thamani kwa asilimia 0.14 tu. Kwa hesabu ya mfano, ili ziada hiyo iishe ingehitajika ongezeko la takriban asilimia 11 la mauzo ya dola, au kupungua kwa asilimia 10 la ununuzi, au ongezeko la asilimia 43 la mauzo kwa Mafuta na Nishati peke yake. Hatari kuu ni kama hali hii itaendelea kwa wiki kadhaa.
Hitimisho: Soko la dola la rejareja linategemea sekta chache: Biashara, Kilimo na Utalii upande wa ugavi; Mafuta na Nishati na Viwanda upande wa mahitaji. Fuatilia ununuzi wa Madini na mauzo ya Mafuta na Nishati kila wiki.
- Retail FX: USD milioni 604.89 hadi 575.68 (−4.8%) · bila Madini: +1.7%
- Madini: 74.18 hadi 35.84 · Mafuta na Nishati: 64.78 hadi 80.55 · Ujenzi: 39.17 hadi 23.42
- Ziada halisi ya dola: 121.35 hadi 29.70 · USD/TZS 2,636.74 (+0.14%)
Chanzo: Benki Kuu ya Tanzania, Domestic Financial Markets Weekly Report (19–23 Oktoba 2026); mahesabu ya TICGL/TERI. Kwa maelezo zaidi: economist@ticgl.com.